Single-Cell RNA-Sequencing Reveals a Continuous Spectrum of Differentiation in Hematopoietic Cells
Identifieur interne : 000346 ( Pmc/Corpus ); précédent : 000345; suivant : 000347Single-Cell RNA-Sequencing Reveals a Continuous Spectrum of Differentiation in Hematopoietic Cells
Auteurs : Iain C. Macaulay ; Valentine Svensson ; Charlotte Labalette ; Lauren Ferreira ; Fiona Hamey ; Thierry Voet ; Sarah A. Teichmann ; Ana CvejicSource :
- Cell Reports [ 2211-1247 ] ; 2016.
Abstract
The transcriptional programs that govern hematopoiesis have been investigated primarily by population-level analysis of hematopoietic stem and progenitor cells, which cannot reveal the continuous nature of the differentiation process. Here we applied single-cell RNA-sequencing to a population of hematopoietic cells in zebrafish as they undergo thrombocyte lineage commitment. By reconstructing their developmental chronology computationally, we were able to place each cell along a continuum from stem cell to mature cell, refining the traditional lineage tree. The progression of cells along this continuum is characterized by a highly coordinated transcriptional program, displaying simultaneous suppression of genes involved in cell proliferation and ribosomal biogenesis as the expression of lineage specific genes increases. Within this program, there is substantial heterogeneity in the expression of the key lineage regulators. Overall, the total number of genes expressed, as well as the total mRNA content of the cell, decreases as the cells undergo lineage commitment.
Url:
DOI: 10.1016/j.celrep.2015.12.082
PubMed: 26804912
PubMed Central: 4742565
Links to Exploration step
PMC:4742565Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en">Single-Cell RNA-Sequencing Reveals a Continuous Spectrum of Differentiation in Hematopoietic Cells</title><author><name sortKey="Macaulay, Iain C" sort="Macaulay, Iain C" uniqKey="Macaulay I" first="Iain C." last="Macaulay">Iain C. Macaulay</name><affiliation><nlm:aff id="aff1">Sanger Institute–EBI Single-Cell Genomics Centre, Wellcome Trust Sanger Institute, Wellcome Trust Genome Campus, Hinxton, Cambridge CB10 1HH, UK</nlm:aff></affiliation></author><author><name sortKey="Svensson, Valentine" sort="Svensson, Valentine" uniqKey="Svensson V" first="Valentine" last="Svensson">Valentine Svensson</name><affiliation><nlm:aff id="aff2">Wellcome Trust Sanger Institute, Wellcome Trust Genome Campus, Hinxton, Cambridge CB10 1HH, UK</nlm:aff></affiliation><affiliation><nlm:aff id="aff3">European Molecular Biology Laboratory, European Bioinformatics Institute (EMBL-EBI), Wellcome Trust Genome Campus, Hinxton, Cambridge CB10 1SD, UK</nlm:aff></affiliation></author><author><name sortKey="Labalette, Charlotte" sort="Labalette, Charlotte" uniqKey="Labalette C" first="Charlotte" last="Labalette">Charlotte Labalette</name><affiliation><nlm:aff id="aff2">Wellcome Trust Sanger Institute, Wellcome Trust Genome Campus, Hinxton, Cambridge CB10 1HH, UK</nlm:aff></affiliation><affiliation><nlm:aff id="aff4">Department of Haematology, University of Cambridge, Cambridge CB2 0PT, UK</nlm:aff></affiliation></author><author><name sortKey="Ferreira, Lauren" sort="Ferreira, Lauren" uniqKey="Ferreira L" first="Lauren" last="Ferreira">Lauren Ferreira</name><affiliation><nlm:aff id="aff2">Wellcome Trust Sanger Institute, Wellcome Trust Genome Campus, Hinxton, Cambridge CB10 1HH, UK</nlm:aff></affiliation><affiliation><nlm:aff id="aff4">Department of Haematology, University of Cambridge, Cambridge CB2 0PT, UK</nlm:aff></affiliation></author><author><name sortKey="Hamey, Fiona" sort="Hamey, Fiona" uniqKey="Hamey F" first="Fiona" last="Hamey">Fiona Hamey</name><affiliation><nlm:aff id="aff5">Wellcome Trust – Medical Research Council Cambridge Stem Cell Institute, Cambridge CB2 1QR, UK</nlm:aff></affiliation></author><author><name sortKey="Voet, Thierry" sort="Voet, Thierry" uniqKey="Voet T" first="Thierry" last="Voet">Thierry Voet</name><affiliation><nlm:aff id="aff1">Sanger Institute–EBI Single-Cell Genomics Centre, Wellcome Trust Sanger Institute, Wellcome Trust Genome Campus, Hinxton, Cambridge CB10 1HH, UK</nlm:aff></affiliation><affiliation><nlm:aff id="aff6">Department of Human Genetics, University of Leuven, Leuven 3000, Belgium</nlm:aff></affiliation></author><author><name sortKey="Teichmann, Sarah A" sort="Teichmann, Sarah A" uniqKey="Teichmann S" first="Sarah A." last="Teichmann">Sarah A. Teichmann</name><affiliation><nlm:aff id="aff2">Wellcome Trust Sanger Institute, Wellcome Trust Genome Campus, Hinxton, Cambridge CB10 1HH, UK</nlm:aff></affiliation><affiliation><nlm:aff id="aff3">European Molecular Biology Laboratory, European Bioinformatics Institute (EMBL-EBI), Wellcome Trust Genome Campus, Hinxton, Cambridge CB10 1SD, UK</nlm:aff></affiliation></author><author><name sortKey="Cvejic, Ana" sort="Cvejic, Ana" uniqKey="Cvejic A" first="Ana" last="Cvejic">Ana Cvejic</name><affiliation><nlm:aff id="aff2">Wellcome Trust Sanger Institute, Wellcome Trust Genome Campus, Hinxton, Cambridge CB10 1HH, UK</nlm:aff></affiliation><affiliation><nlm:aff id="aff4">Department of Haematology, University of Cambridge, Cambridge CB2 0PT, UK</nlm:aff></affiliation><affiliation><nlm:aff id="aff5">Wellcome Trust – Medical Research Council Cambridge Stem Cell Institute, Cambridge CB2 1QR, UK</nlm:aff></affiliation></author></titleStmt><publicationStmt><idno type="wicri:source">PMC</idno><idno type="pmid">26804912</idno><idno type="pmc">4742565</idno><idno type="url">http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4742565</idno><idno type="RBID">PMC:4742565</idno><idno type="doi">10.1016/j.celrep.2015.12.082</idno><date when="2016">2016</date><idno type="wicri:Area/Pmc/Corpus">000346</idno></publicationStmt><sourceDesc><biblStruct><analytic><title xml:lang="en" level="a" type="main">Single-Cell RNA-Sequencing Reveals a Continuous Spectrum of Differentiation in Hematopoietic Cells</title><author><name sortKey="Macaulay, Iain C" sort="Macaulay, Iain C" uniqKey="Macaulay I" first="Iain C." last="Macaulay">Iain C. Macaulay</name><affiliation><nlm:aff id="aff1">Sanger Institute–EBI Single-Cell Genomics Centre, Wellcome Trust Sanger Institute, Wellcome Trust Genome Campus, Hinxton, Cambridge CB10 1HH, UK</nlm:aff></affiliation></author><author><name sortKey="Svensson, Valentine" sort="Svensson, Valentine" uniqKey="Svensson V" first="Valentine" last="Svensson">Valentine Svensson</name><affiliation><nlm:aff id="aff2">Wellcome Trust Sanger Institute, Wellcome Trust Genome Campus, Hinxton, Cambridge CB10 1HH, UK</nlm:aff></affiliation><affiliation><nlm:aff id="aff3">European Molecular Biology Laboratory, European Bioinformatics Institute (EMBL-EBI), Wellcome Trust Genome Campus, Hinxton, Cambridge CB10 1SD, UK</nlm:aff></affiliation></author><author><name sortKey="Labalette, Charlotte" sort="Labalette, Charlotte" uniqKey="Labalette C" first="Charlotte" last="Labalette">Charlotte Labalette</name><affiliation><nlm:aff id="aff2">Wellcome Trust Sanger Institute, Wellcome Trust Genome Campus, Hinxton, Cambridge CB10 1HH, UK</nlm:aff></affiliation><affiliation><nlm:aff id="aff4">Department of Haematology, University of Cambridge, Cambridge CB2 0PT, UK</nlm:aff></affiliation></author><author><name sortKey="Ferreira, Lauren" sort="Ferreira, Lauren" uniqKey="Ferreira L" first="Lauren" last="Ferreira">Lauren Ferreira</name><affiliation><nlm:aff id="aff2">Wellcome Trust Sanger Institute, Wellcome Trust Genome Campus, Hinxton, Cambridge CB10 1HH, UK</nlm:aff></affiliation><affiliation><nlm:aff id="aff4">Department of Haematology, University of Cambridge, Cambridge CB2 0PT, UK</nlm:aff></affiliation></author><author><name sortKey="Hamey, Fiona" sort="Hamey, Fiona" uniqKey="Hamey F" first="Fiona" last="Hamey">Fiona Hamey</name><affiliation><nlm:aff id="aff5">Wellcome Trust – Medical Research Council Cambridge Stem Cell Institute, Cambridge CB2 1QR, UK</nlm:aff></affiliation></author><author><name sortKey="Voet, Thierry" sort="Voet, Thierry" uniqKey="Voet T" first="Thierry" last="Voet">Thierry Voet</name><affiliation><nlm:aff id="aff1">Sanger Institute–EBI Single-Cell Genomics Centre, Wellcome Trust Sanger Institute, Wellcome Trust Genome Campus, Hinxton, Cambridge CB10 1HH, UK</nlm:aff></affiliation><affiliation><nlm:aff id="aff6">Department of Human Genetics, University of Leuven, Leuven 3000, Belgium</nlm:aff></affiliation></author><author><name sortKey="Teichmann, Sarah A" sort="Teichmann, Sarah A" uniqKey="Teichmann S" first="Sarah A." last="Teichmann">Sarah A. Teichmann</name><affiliation><nlm:aff id="aff2">Wellcome Trust Sanger Institute, Wellcome Trust Genome Campus, Hinxton, Cambridge CB10 1HH, UK</nlm:aff></affiliation><affiliation><nlm:aff id="aff3">European Molecular Biology Laboratory, European Bioinformatics Institute (EMBL-EBI), Wellcome Trust Genome Campus, Hinxton, Cambridge CB10 1SD, UK</nlm:aff></affiliation></author><author><name sortKey="Cvejic, Ana" sort="Cvejic, Ana" uniqKey="Cvejic A" first="Ana" last="Cvejic">Ana Cvejic</name><affiliation><nlm:aff id="aff2">Wellcome Trust Sanger Institute, Wellcome Trust Genome Campus, Hinxton, Cambridge CB10 1HH, UK</nlm:aff></affiliation><affiliation><nlm:aff id="aff4">Department of Haematology, University of Cambridge, Cambridge CB2 0PT, UK</nlm:aff></affiliation><affiliation><nlm:aff id="aff5">Wellcome Trust – Medical Research Council Cambridge Stem Cell Institute, Cambridge CB2 1QR, UK</nlm:aff></affiliation></author></analytic><series><title level="j">Cell Reports</title><idno type="eISSN">2211-1247</idno><imprint><date when="2016">2016</date></imprint></series></biblStruct></sourceDesc></fileDesc><profileDesc><textClass></textClass></profileDesc></teiHeader><front><div type="abstract" xml:lang="en"><title>Summary</title><p>The transcriptional programs that govern hematopoiesis have been investigated primarily by population-level analysis of hematopoietic stem and progenitor cells, which cannot reveal the continuous nature of the differentiation process. Here we applied single-cell RNA-sequencing to a population of hematopoietic cells in zebrafish as they undergo thrombocyte lineage commitment. By reconstructing their developmental chronology computationally, we were able to place each cell along a continuum from stem cell to mature cell, refining the traditional lineage tree. The progression of cells along this continuum is characterized by a highly coordinated transcriptional program, displaying simultaneous suppression of genes involved in cell proliferation and ribosomal biogenesis as the expression of lineage specific genes increases. Within this program, there is substantial heterogeneity in the expression of the key lineage regulators. Overall, the total number of genes expressed, as well as the total mRNA content of the cell, decreases as the cells undergo lineage commitment.</p></div></front><back><div1 type="bibliography"><listBibl><biblStruct><analytic><author><name sortKey="Bielczyk Maczy Ska, E" uniqKey="Bielczyk Maczy Ska E">E. Bielczyk-Maczyńska</name></author><author><name sortKey="Serbanovic Canic, J" uniqKey="Serbanovic Canic J">J. Serbanovic-Canic</name></author><author><name sortKey="Ferreira, L" uniqKey="Ferreira L">L. Ferreira</name></author><author><name sortKey="Soranzo, N" uniqKey="Soranzo N">N. Soranzo</name></author><author><name sortKey="Stemple, D L" uniqKey="Stemple D">D.L. Stemple</name></author><author><name sortKey="Ouwehand, W H" uniqKey="Ouwehand W">W.H. Ouwehand</name></author><author><name sortKey="Cvejic, A" uniqKey="Cvejic A">A. Cvejic</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Capron, C" uniqKey="Capron C">C. Capron</name></author><author><name sortKey="Lecluse, Y" uniqKey="Lecluse Y">Y. Lécluse</name></author><author><name sortKey="Kaushik, A L" uniqKey="Kaushik A">A.L. Kaushik</name></author><author><name sortKey="Foudi, A" uniqKey="Foudi A">A. Foudi</name></author><author><name sortKey="Lacout, C" uniqKey="Lacout C">C. Lacout</name></author><author><name sortKey="Sekkai, D" uniqKey="Sekkai D">D. Sekkai</name></author><author><name sortKey="Godin, I" uniqKey="Godin I">I. Godin</name></author><author><name sortKey="Albagli, O" uniqKey="Albagli O">O. Albagli</name></author><author><name sortKey="Poullion, I" uniqKey="Poullion I">I. Poullion</name></author><author><name sortKey="Svinartchouk, F" uniqKey="Svinartchouk F">F. Svinartchouk</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Carradice, D" uniqKey="Carradice D">D. Carradice</name></author><author><name sortKey="Lieschke, G J" uniqKey="Lieschke G">G.J. Lieschke</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Carrillo, M" uniqKey="Carrillo M">M. Carrillo</name></author><author><name sortKey="Kim, S" uniqKey="Kim S">S. Kim</name></author><author><name sortKey="Rajpurohit, S K" uniqKey="Rajpurohit S">S.K. Rajpurohit</name></author><author><name sortKey="Kulkarni, V" uniqKey="Kulkarni V">V. Kulkarni</name></author><author><name sortKey="Jagadeeswaran, P" uniqKey="Jagadeeswaran P">P. Jagadeeswaran</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Clay, D" uniqKey="Clay D">D. Clay</name></author><author><name sortKey="Rubinstein, E" uniqKey="Rubinstein E">E. Rubinstein</name></author><author><name sortKey="Mishal, Z" uniqKey="Mishal Z">Z. Mishal</name></author><author><name sortKey="Anjo, A" uniqKey="Anjo A">A. Anjo</name></author><author><name sortKey="Prenant, M" uniqKey="Prenant M">M. Prenant</name></author><author><name sortKey="Jasmin, C" uniqKey="Jasmin C">C. Jasmin</name></author><author><name sortKey="Boucheix, C" uniqKey="Boucheix C">C. Boucheix</name></author><author><name sortKey="Le Bousse Kerdiles, M C" uniqKey="Le Bousse Kerdiles M">M.C. Le Bousse-Kerdilès</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Cvejic, A" uniqKey="Cvejic A">A. Cvejic</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Debili, N" uniqKey="Debili N">N. Debili</name></author><author><name sortKey="Robin, C" uniqKey="Robin C">C. Robin</name></author><author><name sortKey="Schiavon, V" uniqKey="Schiavon V">V. Schiavon</name></author><author><name sortKey="Letestu, R" uniqKey="Letestu R">R. Letestu</name></author><author><name sortKey="Pflumio, F" uniqKey="Pflumio F">F. Pflumio</name></author><author><name sortKey="Mitjavila Garcia, M T" uniqKey="Mitjavila Garcia M">M.T. Mitjavila-Garcia</name></author><author><name sortKey="Coulombel, L" uniqKey="Coulombel L">L. Coulombel</name></author><author><name sortKey="Vainchenker, W" uniqKey="Vainchenker W">W. Vainchenker</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Deng, Q" uniqKey="Deng Q">Q. Deng</name></author><author><name sortKey="Wang, Q" uniqKey="Wang Q">Q. Wang</name></author><author><name sortKey="Zong, W Y" uniqKey="Zong W">W.-Y. Zong</name></author><author><name sortKey="Zheng, D L" uniqKey="Zheng D">D.-L. Zheng</name></author><author><name sortKey="Wen, Y X" uniqKey="Wen Y">Y.-X. Wen</name></author><author><name sortKey="Wang, K S" uniqKey="Wang K">K.-S. Wang</name></author><author><name sortKey="Teng, X M" uniqKey="Teng X">X.-M. Teng</name></author><author><name sortKey="Zhang, X" uniqKey="Zhang X">X. Zhang</name></author><author><name sortKey="Huang, J" uniqKey="Huang J">J. Huang</name></author><author><name sortKey="Han, Z G" uniqKey="Han Z">Z.-G. Han</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Dobin, A" uniqKey="Dobin A">A. Dobin</name></author><author><name sortKey="Davis, C A" uniqKey="Davis C">C.A. Davis</name></author><author><name sortKey="Schlesinger, F" uniqKey="Schlesinger F">F. Schlesinger</name></author><author><name sortKey="Drenkow, J" uniqKey="Drenkow J">J. Drenkow</name></author><author><name sortKey="Zaleski, C" uniqKey="Zaleski C">C. Zaleski</name></author><author><name sortKey="Jha, S" uniqKey="Jha S">S. Jha</name></author><author><name sortKey="Batut, P" uniqKey="Batut P">P. Batut</name></author><author><name sortKey="Chaisson, M" uniqKey="Chaisson M">M. Chaisson</name></author><author><name sortKey="Gingeras, T R" uniqKey="Gingeras T">T.R. Gingeras</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Downes, C S" uniqKey="Downes C">C.S. Downes</name></author><author><name sortKey="Clarke, D J" uniqKey="Clarke D">D.J. Clarke</name></author><author><name sortKey="Mullinger, A M" uniqKey="Mullinger A">A.M. Mullinger</name></author><author><name sortKey="Gimenez Abian, J F" uniqKey="Gimenez Abian J">J.F. Giménez-Abián</name></author><author><name sortKey="Creighton, A M" uniqKey="Creighton A">A.M. Creighton</name></author><author><name sortKey="Johnson, R T" uniqKey="Johnson R">R.T. Johnson</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Geurts, P" uniqKey="Geurts P">P. Geurts</name></author><author><name sortKey="Ernst, D" uniqKey="Ernst D">D. Ernst</name></author><author><name sortKey="Wehenkel, L" uniqKey="Wehenkel L">L. Wehenkel</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Greig, K T" uniqKey="Greig K">K.T. Greig</name></author><author><name sortKey="Carotta, S" uniqKey="Carotta S">S. Carotta</name></author><author><name sortKey="Nutt, S L" uniqKey="Nutt S">S.L. Nutt</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Guo, G" uniqKey="Guo G">G. Guo</name></author><author><name sortKey="Luc, S" uniqKey="Luc S">S. Luc</name></author><author><name sortKey="Marco, E" uniqKey="Marco E">E. Marco</name></author><author><name sortKey="Lin, T W" uniqKey="Lin T">T.-W. Lin</name></author><author><name sortKey="Peng, C" uniqKey="Peng C">C. Peng</name></author><author><name sortKey="Kerenyi, M A" uniqKey="Kerenyi M">M.A. Kerenyi</name></author><author><name sortKey="Beyaz, S" uniqKey="Beyaz S">S. Beyaz</name></author><author><name sortKey="Kim, W" uniqKey="Kim W">W. Kim</name></author><author><name sortKey="Xu, J" uniqKey="Xu J">J. Xu</name></author><author><name sortKey="Das, P P" uniqKey="Das P">P.P. Das</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Hensman, J" uniqKey="Hensman J">J. Hensman</name></author><author><name sortKey="Rattray, M" uniqKey="Rattray M">M. Rattray</name></author><author><name sortKey="Lawrence, N D" uniqKey="Lawrence N">N.D. Lawrence</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Howe, K" uniqKey="Howe K">K. Howe</name></author><author><name sortKey="Clark, M D" uniqKey="Clark M">M.D. Clark</name></author><author><name sortKey="Torroja, C F" uniqKey="Torroja C">C.F. Torroja</name></author><author><name sortKey="Torrance, J" uniqKey="Torrance J">J. Torrance</name></author><author><name sortKey="Berthelot, C" uniqKey="Berthelot C">C. Berthelot</name></author><author><name sortKey="Muffato, M" uniqKey="Muffato M">M. Muffato</name></author><author><name sortKey="Collins, J E" uniqKey="Collins J">J.E. Collins</name></author><author><name sortKey="Humphray, S" uniqKey="Humphray S">S. Humphray</name></author><author><name sortKey="Mclaren, K" uniqKey="Mclaren K">K. McLaren</name></author><author><name sortKey="Matthews, L" uniqKey="Matthews L">L. Matthews</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Hsia, N" uniqKey="Hsia N">N. Hsia</name></author><author><name sortKey="Zon, L I" uniqKey="Zon L">L.I. Zon</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Hyv Rinen, A" uniqKey="Hyv Rinen A">A. Hyvärinen</name></author><author><name sortKey="Oja, E" uniqKey="Oja E">E. Oja</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Jagannathan Bogdan, M" uniqKey="Jagannathan Bogdan M">M. Jagannathan-Bogdan</name></author><author><name sortKey="Zon, L I" uniqKey="Zon L">L.I. Zon</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Li, B" uniqKey="Li B">B. Li</name></author><author><name sortKey="Dewey, C N" uniqKey="Dewey C">C.N. Dewey</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Loughran, S J" uniqKey="Loughran S">S.J. Loughran</name></author><author><name sortKey="Kruse, E A" uniqKey="Kruse E">E.A. Kruse</name></author><author><name sortKey="Hacking, D F" uniqKey="Hacking D">D.F. Hacking</name></author><author><name sortKey="De Graaf, C A" uniqKey="De Graaf C">C.A. de Graaf</name></author><author><name sortKey="Hyland, C D" uniqKey="Hyland C">C.D. Hyland</name></author><author><name sortKey="Willson, T A" uniqKey="Willson T">T.A. Willson</name></author><author><name sortKey="Henley, K J" uniqKey="Henley K">K.J. Henley</name></author><author><name sortKey="Ellis, S" uniqKey="Ellis S">S. Ellis</name></author><author><name sortKey="Voss, A K" uniqKey="Voss A">A.K. Voss</name></author><author><name sortKey="Metcalf, D" uniqKey="Metcalf D">D. Metcalf</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Ma, D" uniqKey="Ma D">D. Ma</name></author><author><name sortKey="Zhang, J" uniqKey="Zhang J">J. Zhang</name></author><author><name sortKey="Lin, H F" uniqKey="Lin H">H.-F. Lin</name></author><author><name sortKey="Italiano, J" uniqKey="Italiano J">J. Italiano</name></author><author><name sortKey="Handin, R I" uniqKey="Handin R">R.I. Handin</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Meyer, A" uniqKey="Meyer A">A. Meyer</name></author><author><name sortKey="Schartl, M" uniqKey="Schartl M">M. Schartl</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Muller Sieburg, C E" uniqKey="Muller Sieburg C">C.E. Muller-Sieburg</name></author><author><name sortKey="Sieburg, H B" uniqKey="Sieburg H">H.B. Sieburg</name></author><author><name sortKey="Bernitz, J M" uniqKey="Bernitz J">J.M. Bernitz</name></author><author><name sortKey="Cattarossi, G" uniqKey="Cattarossi G">G. Cattarossi</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Notta, F" uniqKey="Notta F">F. Notta</name></author><author><name sortKey="Zandi, S" uniqKey="Zandi S">S. Zandi</name></author><author><name sortKey="Takayama, N" uniqKey="Takayama N">N. Takayama</name></author><author><name sortKey="Dobson, S" uniqKey="Dobson S">S. Dobson</name></author><author><name sortKey="Gan, O I" uniqKey="Gan O">O.I. Gan</name></author><author><name sortKey="Wilson, G" uniqKey="Wilson G">G. Wilson</name></author><author><name sortKey="Kaufmann, K B" uniqKey="Kaufmann K">K.B. Kaufmann</name></author><author><name sortKey="Mcleod, J" uniqKey="Mcleod J">J. McLeod</name></author><author><name sortKey="Laurenti, E" uniqKey="Laurenti E">E. Laurenti</name></author><author><name sortKey="Dunant, C F" uniqKey="Dunant C">C.F. Dunant</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Orkin, S H" uniqKey="Orkin S">S.H. Orkin</name></author><author><name sortKey="Zon, L I" uniqKey="Zon L">L.I. Zon</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Patro, R" uniqKey="Patro R">R. Patro</name></author><author><name sortKey="Mount, S M" uniqKey="Mount S">S.M. Mount</name></author><author><name sortKey="Kingsford, C" uniqKey="Kingsford C">C. Kingsford</name></author></analytic></biblStruct><biblStruct></biblStruct><biblStruct><analytic><author><name sortKey="Pedregosa, F" uniqKey="Pedregosa F">F. Pedregosa</name></author><author><name sortKey="Varoquaux, G" uniqKey="Varoquaux G">G. Varoquaux</name></author><author><name sortKey="Gramfort, A" uniqKey="Gramfort A">A. Gramfort</name></author><author><name sortKey="Michel, V" uniqKey="Michel V">V. Michel</name></author><author><name sortKey="Thirion, B" uniqKey="Thirion B">B. Thirion</name></author><author><name sortKey="Grisel, O" uniqKey="Grisel O">O. Grisel</name></author><author><name sortKey="Blondel, M" uniqKey="Blondel M">M. Blondel</name></author><author><name sortKey="Prettenhofer, P" uniqKey="Prettenhofer P">P. Prettenhofer</name></author><author><name sortKey="Weiss, R" uniqKey="Weiss R">R. Weiss</name></author><author><name sortKey="Dubourg, V" uniqKey="Dubourg V">V. Dubourg</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Picelli, S" uniqKey="Picelli S">S. Picelli</name></author><author><name sortKey="Bjorklund, K" uniqKey="Bjorklund ">Å.K. Björklund</name></author><author><name sortKey="Faridani, O R" uniqKey="Faridani O">O.R. Faridani</name></author><author><name sortKey="Sagasser, S" uniqKey="Sagasser S">S. Sagasser</name></author><author><name sortKey="Winberg, G" uniqKey="Winberg G">G. Winberg</name></author><author><name sortKey="Sandberg, R" uniqKey="Sandberg R">R. Sandberg</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Picelli, S" uniqKey="Picelli S">S. Picelli</name></author><author><name sortKey="Faridani, O R" uniqKey="Faridani O">O.R. Faridani</name></author><author><name sortKey="Bjorklund, A K" uniqKey="Bjorklund A">A.K. Björklund</name></author><author><name sortKey="Winberg, G" uniqKey="Winberg G">G. Winberg</name></author><author><name sortKey="Sagasser, S" uniqKey="Sagasser S">S. Sagasser</name></author><author><name sortKey="Sandberg, R" uniqKey="Sandberg R">R. Sandberg</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Pineault, N" uniqKey="Pineault N">N. Pineault</name></author><author><name sortKey="Helgason, C D" uniqKey="Helgason C">C.D. Helgason</name></author><author><name sortKey="Lawrence, H J" uniqKey="Lawrence H">H.J. Lawrence</name></author><author><name sortKey="Humphries, R K" uniqKey="Humphries R">R.K. Humphries</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Poirault Chassac, S" uniqKey="Poirault Chassac S">S. Poirault-Chassac</name></author><author><name sortKey="Six, E" uniqKey="Six E">E. Six</name></author><author><name sortKey="Catelain, C" uniqKey="Catelain C">C. Catelain</name></author><author><name sortKey="Lavergne, M" uniqKey="Lavergne M">M. Lavergne</name></author><author><name sortKey="Villeval, J L" uniqKey="Villeval J">J.-L. Villeval</name></author><author><name sortKey="Vainchenker, W" uniqKey="Vainchenker W">W. Vainchenker</name></author><author><name sortKey="Lauret, E" uniqKey="Lauret E">E. Lauret</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Reimand, J" uniqKey="Reimand J">J. Reimand</name></author><author><name sortKey="Arak, T" uniqKey="Arak T">T. Arak</name></author><author><name sortKey="Vilo, J" uniqKey="Vilo J">J. Vilo</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Robin, C" uniqKey="Robin C">C. Robin</name></author><author><name sortKey="Ottersbach, K" uniqKey="Ottersbach K">K. Ottersbach</name></author><author><name sortKey="Boisset, J C" uniqKey="Boisset J">J.-C. Boisset</name></author><author><name sortKey="Oziemlak, A" uniqKey="Oziemlak A">A. Oziemlak</name></author><author><name sortKey="Dzierzak, E" uniqKey="Dzierzak E">E. Dzierzak</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Sanjuan Pla, A" uniqKey="Sanjuan Pla A">A. Sanjuan-Pla</name></author><author><name sortKey="Macaulay, I C" uniqKey="Macaulay I">I.C. Macaulay</name></author><author><name sortKey="Jensen, C T" uniqKey="Jensen C">C.T. Jensen</name></author><author><name sortKey="Woll, P S" uniqKey="Woll P">P.S. Woll</name></author><author><name sortKey="Luis, T C" uniqKey="Luis T">T.C. Luis</name></author><author><name sortKey="Mead, A" uniqKey="Mead A">A. Mead</name></author><author><name sortKey="Moore, S" uniqKey="Moore S">S. Moore</name></author><author><name sortKey="Carella, C" uniqKey="Carella C">C. Carella</name></author><author><name sortKey="Matsuoka, S" uniqKey="Matsuoka S">S. Matsuoka</name></author><author><name sortKey="Bouriez Jones, T" uniqKey="Bouriez Jones T">T. Bouriez Jones</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Schick, P K" uniqKey="Schick P">P.K. Schick</name></author><author><name sortKey="Konkle, B A" uniqKey="Konkle B">B.A. Konkle</name></author><author><name sortKey="He, X" uniqKey="He X">X. He</name></author><author><name sortKey="Thornton, R D" uniqKey="Thornton R">R.D. Thornton</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Schulte, R" uniqKey="Schulte R">R. Schulte</name></author><author><name sortKey="Wilson, N K" uniqKey="Wilson N">N.K. Wilson</name></author><author><name sortKey="Prick, J C M" uniqKey="Prick J">J.C.M. Prick</name></author><author><name sortKey="Cossetti, C" uniqKey="Cossetti C">C. Cossetti</name></author><author><name sortKey="Maj, M K" uniqKey="Maj M">M.K. Maj</name></author><author><name sortKey="Gottgens, B" uniqKey="Gottgens B">B. Gottgens</name></author><author><name sortKey="Kent, D G" uniqKey="Kent D">D.G. Kent</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Seita, J" uniqKey="Seita J">J. Seita</name></author><author><name sortKey="Weissman, I L" uniqKey="Weissman I">I.L. Weissman</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Song, H D" uniqKey="Song H">H.-D. Song</name></author><author><name sortKey="Sun, X J" uniqKey="Sun X">X.-J. Sun</name></author><author><name sortKey="Deng, M" uniqKey="Deng M">M. Deng</name></author><author><name sortKey="Zhang, G W" uniqKey="Zhang G">G.-W. Zhang</name></author><author><name sortKey="Zhou, Y" uniqKey="Zhou Y">Y. Zhou</name></author><author><name sortKey="Wu, X Y" uniqKey="Wu X">X.-Y. Wu</name></author><author><name sortKey="Sheng, Y" uniqKey="Sheng Y">Y. Sheng</name></author><author><name sortKey="Chen, Y" uniqKey="Chen Y">Y. Chen</name></author><author><name sortKey="Ruan, Z" uniqKey="Ruan Z">Z. Ruan</name></author><author><name sortKey="Jiang, C L" uniqKey="Jiang C">C.-L. Jiang</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Stachura, D L" uniqKey="Stachura D">D.L. Stachura</name></author><author><name sortKey="Reyes, J R" uniqKey="Reyes J">J.R. Reyes</name></author><author><name sortKey="Bartunek, P" uniqKey="Bartunek P">P. Bartunek</name></author><author><name sortKey="Paw, B H" uniqKey="Paw B">B.H. Paw</name></author><author><name sortKey="Zon, L I" uniqKey="Zon L">L.I. Zon</name></author><author><name sortKey="Traver, D" uniqKey="Traver D">D. Traver</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Tenen, D G" uniqKey="Tenen D">D.G. Tenen</name></author><author><name sortKey="Hromas, R" uniqKey="Hromas R">R. Hromas</name></author><author><name sortKey="Licht, J D" uniqKey="Licht J">J.D. Licht</name></author><author><name sortKey="Zhang, D E" uniqKey="Zhang D">D.E. Zhang</name></author></analytic></biblStruct><biblStruct></biblStruct><biblStruct></biblStruct><biblStruct><analytic><author><name sortKey="Van Der Maaten, L" uniqKey="Van Der Maaten L">L. Van der Maaten</name></author><author><name sortKey="Hinton, G" uniqKey="Hinton G">G. Hinton</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Vassen, L" uniqKey="Vassen L">L. Vassen</name></author><author><name sortKey="Okayama, T" uniqKey="Okayama T">T. Okayama</name></author><author><name sortKey="Moroy, T" uniqKey="Moroy T">T. Möröy</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Wagner, G P" uniqKey="Wagner G">G.P. Wagner</name></author><author><name sortKey="Kin, K" uniqKey="Kin K">K. Kin</name></author><author><name sortKey="Lynch, V J" uniqKey="Lynch V">V.J. Lynch</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Wang, L" uniqKey="Wang L">L. Wang</name></author><author><name sortKey="Wang, S" uniqKey="Wang S">S. Wang</name></author><author><name sortKey="Li, W" uniqKey="Li W">W. Li</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Ward, J H" uniqKey="Ward J">J.H. Ward</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Wright, D E" uniqKey="Wright D">D.E. Wright</name></author><author><name sortKey="Bowman, E P" uniqKey="Bowman E">E.P. Bowman</name></author><author><name sortKey="Wagers, A J" uniqKey="Wagers A">A.J. Wagers</name></author><author><name sortKey="Butcher, E C" uniqKey="Butcher E">E.C. Butcher</name></author><author><name sortKey="Weissman, I L" uniqKey="Weissman I">I.L. Weissman</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Zeng, H" uniqKey="Zeng H">H. Zeng</name></author><author><name sortKey="Yucel, R" uniqKey="Yucel R">R. Yücel</name></author><author><name sortKey="Kosan, C" uniqKey="Kosan C">C. Kosan</name></author><author><name sortKey="Klein Hitpass, L" uniqKey="Klein Hitpass L">L. Klein-Hitpass</name></author><author><name sortKey="Moroy, T" uniqKey="Moroy T">T. Möröy</name></author></analytic></biblStruct></listBibl></div1></back></TEI><pmc article-type="research-article"><pmc-dir>properties open_access</pmc-dir>
<front><journal-meta><journal-id journal-id-type="nlm-ta">Cell Rep</journal-id><journal-id journal-id-type="iso-abbrev">Cell Rep</journal-id><journal-title-group><journal-title>Cell Reports</journal-title></journal-title-group><issn pub-type="epub">2211-1247</issn><publisher><publisher-name>Cell Press</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="pmid">26804912</article-id><article-id pub-id-type="pmc">4742565</article-id><article-id pub-id-type="publisher-id">S2211-1247(15)01538-7</article-id><article-id pub-id-type="doi">10.1016/j.celrep.2015.12.082</article-id><article-categories><subj-group subj-group-type="heading"><subject>Resource</subject></subj-group></article-categories><title-group><article-title>Single-Cell RNA-Sequencing Reveals a Continuous Spectrum of Differentiation in Hematopoietic Cells</article-title></title-group><contrib-group><contrib contrib-type="author"><name><surname>Macaulay</surname><given-names>Iain C.</given-names></name><xref rid="aff1" ref-type="aff">1</xref><xref rid="fn1" ref-type="fn">7</xref></contrib><contrib contrib-type="author"><name><surname>Svensson</surname><given-names>Valentine</given-names></name><xref rid="aff2" ref-type="aff">2</xref><xref rid="aff3" ref-type="aff">3</xref><xref rid="fn1" ref-type="fn">7</xref></contrib><contrib contrib-type="author"><name><surname>Labalette</surname><given-names>Charlotte</given-names></name><xref rid="aff2" ref-type="aff">2</xref><xref rid="aff4" ref-type="aff">4</xref><xref rid="fn1" ref-type="fn">7</xref></contrib><contrib contrib-type="author"><name><surname>Ferreira</surname><given-names>Lauren</given-names></name><xref rid="aff2" ref-type="aff">2</xref><xref rid="aff4" ref-type="aff">4</xref></contrib><contrib contrib-type="author"><name><surname>Hamey</surname><given-names>Fiona</given-names></name><xref rid="aff5" ref-type="aff">5</xref></contrib><contrib contrib-type="author"><name><surname>Voet</surname><given-names>Thierry</given-names></name><xref rid="aff1" ref-type="aff">1</xref><xref rid="aff6" ref-type="aff">6</xref></contrib><contrib contrib-type="author"><name><surname>Teichmann</surname><given-names>Sarah A.</given-names></name><xref rid="aff2" ref-type="aff">2</xref><xref rid="aff3" ref-type="aff">3</xref><xref rid="fn2" ref-type="fn">8</xref></contrib><contrib contrib-type="author"><name><surname>Cvejic</surname><given-names>Ana</given-names></name><email>as889@cam.ac.uk</email><xref rid="aff2" ref-type="aff">2</xref><xref rid="aff4" ref-type="aff">4</xref><xref rid="aff5" ref-type="aff">5</xref><xref rid="fn2" ref-type="fn">8</xref><xref rid="cor1" ref-type="corresp">∗</xref></contrib></contrib-group><aff id="aff1"><label>1</label>Sanger Institute–EBI Single-Cell Genomics Centre, Wellcome Trust Sanger Institute, Wellcome Trust Genome Campus, Hinxton, Cambridge CB10 1HH, UK</aff><aff id="aff2"><label>2</label>Wellcome Trust Sanger Institute, Wellcome Trust Genome Campus, Hinxton, Cambridge CB10 1HH, UK</aff><aff id="aff3"><label>3</label>European Molecular Biology Laboratory, European Bioinformatics Institute (EMBL-EBI), Wellcome Trust Genome Campus, Hinxton, Cambridge CB10 1SD, UK</aff><aff id="aff4"><label>4</label>Department of Haematology, University of Cambridge, Cambridge CB2 0PT, UK</aff><aff id="aff5"><label>5</label>Wellcome Trust – Medical Research Council Cambridge Stem Cell Institute, Cambridge CB2 1QR, UK</aff><aff id="aff6"><label>6</label>Department of Human Genetics, University of Leuven, Leuven 3000, Belgium</aff><author-notes><corresp id="cor1"><label>∗</label>Corresponding author <email>as889@cam.ac.uk</email></corresp><fn id="fn1"><label>7</label><p id="ntpara0010">Co-first author</p></fn><fn id="fn2"><label>8</label><p id="ntpara0015">Co-senior author</p></fn></author-notes><pub-date pub-type="pmc-release"><day>21</day><month>1</month><year>2016</year></pub-date><pmc-comment> PMC Release delay is 0 months and 0 days and was based on .</pmc-comment>
<pub-date pub-type="collection"><day>02</day><month>2</month><year>2016</year></pub-date><pub-date pub-type="epub"><day>21</day><month>1</month><year>2016</year></pub-date><volume>14</volume><issue>4</issue><fpage>966</fpage><lpage>977</lpage><history><date date-type="received"><day>10</day><month>8</month><year>2015</year></date><date date-type="rev-recd"><day>30</day><month>10</month><year>2015</year></date><date date-type="accepted"><day>16</day><month>12</month><year>2015</year></date></history><permissions><copyright-statement>© 2016 The Authors</copyright-statement><copyright-year>2016</copyright-year><license license-type="CC BY" xlink:href="http://creativecommons.org/licenses/by/4.0/"><license-p>This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).</license-p></license></permissions><abstract><title>Summary</title><p>The transcriptional programs that govern hematopoiesis have been investigated primarily by population-level analysis of hematopoietic stem and progenitor cells, which cannot reveal the continuous nature of the differentiation process. Here we applied single-cell RNA-sequencing to a population of hematopoietic cells in zebrafish as they undergo thrombocyte lineage commitment. By reconstructing their developmental chronology computationally, we were able to place each cell along a continuum from stem cell to mature cell, refining the traditional lineage tree. The progression of cells along this continuum is characterized by a highly coordinated transcriptional program, displaying simultaneous suppression of genes involved in cell proliferation and ribosomal biogenesis as the expression of lineage specific genes increases. Within this program, there is substantial heterogeneity in the expression of the key lineage regulators. Overall, the total number of genes expressed, as well as the total mRNA content of the cell, decreases as the cells undergo lineage commitment.</p></abstract><abstract abstract-type="graphical"><title>Graphical Abstract</title><fig id="undfig1" position="anchor"><graphic xlink:href="fx1"></graphic></fig></abstract><abstract abstract-type="author-highlights"><title>Highlights</title><p><list list-type="simple"><list-item id="u0010"><label>•</label><p>Single-cell RNA-sequencing reveals the continuous nature of thrombocyte development</p></list-item><list-item id="u0015"><label>•</label><p>Coordinated transcriptional programs govern progression of differentiation</p></list-item><list-item id="u0020"><label>•</label><p>Number of genes expressed and mRNA content per cell decrease during differentiation</p></list-item><list-item id="u0025"><label>•</label><p>Zebrafish thrombocytes remain transcriptionally active in circulation</p></list-item></list></p></abstract><abstract abstract-type="teaser"><p>Computational reconstruction of the thrombocyte’s developmental chronology from scRNA-seq data reveals the continuous nature of the differentiation process. Macaulay et al. show that a highly coordinated transcriptional program characterizes the progression of cells along this continuum. Within this program, there is substantial heterogeneity in the expression of key lineage regulators.</p></abstract></article-meta><notes><p id="misc0010">Published: January 21, 2016</p></notes></front><body><sec id="sec1"><title>Introduction</title><p>Hematopoietic stem cells (HSCs) have the ability to self-renew and produce cells that give rise to all different blood cell types (<xref rid="bib25" ref-type="bibr">Orkin and Zon, 2008</xref>). Our understanding of the functional properties of these various hematopoietic cell types has been advanced mainly by population level analysis. Current methods of purifying hematopoietic cells to relative homogeneity are based on the expression of specific combinations of cell surface markers. However, a homogeneous population of cells, as determined by a well-defined set of cell surface markers, may include many functionally distinct populations. This was nicely illustrated in studies showing that within the HSC compartment, individual HSCs may have different reconstitution patterns (e.g., balanced production of myeloid and lymphoid cells or deficiency in lymphoid potential) (<xref rid="bib23" ref-type="bibr">Muller-Sieburg et al., 2012</xref>). More recently, it was demonstrated that common myeloid progenitors (CMP) are a mixed population of cells with distinct lineage potentials (<xref rid="bib24" ref-type="bibr">Notta et al., 2015</xref>). The lack of CMPs as a separate cell entity with broad myeloid potential brings into question the traditional model of hematopoietic lineage development and further underscores the importance of revising the current view of lineage development in hematopoiesis. Therefore, there is a need to address the exact composition of the stem and progenitor populations in vivo, as well as the relationships between them. Single cell transcriptome analysis might provide answers to these outstanding questions (<xref rid="bib6" ref-type="bibr">Cvejic, 2015</xref>).</p><p>Among vertebrate models, the zebrafish provides a unique combination of advantages for the study of blood development at the single cell level. Zebrafish blood contains cells of all hematopoietic lineages and orthologs of most transcription factors involved in mammalian hematopoiesis (<xref rid="bib16" ref-type="bibr">Hsia and Zon, 2005</xref>, <xref rid="bib39" ref-type="bibr">Song et al., 2004</xref>). Importantly, transcriptional mechanisms and signaling pathways in hematopoiesis are well conserved between zebrafish and mammals, making them a clinically relevant model system (<xref rid="bib18" ref-type="bibr">Jagannathan-Bogdan and Zon, 2013</xref>).</p><p>Over the past few years, a number of transgenic zebrafish lines were generated in which hematopoietic cell specific promoters drive expression of fluorescent molecules (<xref rid="bib3" ref-type="bibr">Carradice and Lieschke, 2008</xref>). These reporter lines provide a valuable resource of labeled cells ranging from HSCs to a wide range of mature blood cell types. As in mammals, adult hematopoiesis in zebrafish is both continuous and asynchronous. Thus, a single sample of kidney marrow (the analogous tissue to mammalian bone marrow) contains the full spectrum of hematopoietic cell types at various stages of differentiation at any one time. As this is the single site of hematopoiesis in zebrafish, and is easily accessible, the cells are minimally perturbed when sorted ex vivo, making this an ideal system to study basic principles of regulation of differentiation, both at the molecular and cellular levels.</p><p>Here we used high-throughput single-cell RNA sequencing combined with fluorescence-activated cell sorting index sorting analysis of adult zebrafish marrow-derived hematopoietic cells. We ordered cells by their progression through differentiation based on gene expression profiles using no prior knowledge of which cell population they belong to, as defined by surface markers. Our analysis revealed the continuous nature of thrombocyte development and the coordinated transcriptional programs that govern differentiation progression. Interestingly, thrombocytes in zebrafish remain transcriptionally active even after leaving the kidney marrow and entering the circulation.</p></sec><sec id="sec2"><title>Results</title><sec id="sec2.1"><title>Profiling Individual Hematopoietic Cells Ex Vivo</title><p>Here, we used single-cell RNA-sequencing (RNA-seq) of zebrafish kidney cells to resolve the cellular hierarchy of lineage development in the myeloid branch of hematopoiesis. To focus on this lineage, we used expression of CD41 as a marker of HSCs and the megakaryocyte equivalent in fish (“thrombocytes”). CD41 in human is highly regulated during hematopoietic development (<xref rid="bib7" ref-type="bibr">Debili et al., 2001</xref>, <xref rid="bib34" ref-type="bibr">Robin et al., 2011</xref>), and in zebrafish, the <italic>Tg(cd41:EGFP)</italic> reporter line labels two distinct populations of cells that express the cd41-EGFP transgene. The weakly fluorescent (EGFP<sup>low</sup>) subset marks HSCs and progenitor cells (<xref rid="bib21" ref-type="bibr">Ma et al., 2011</xref>), and the brightly fluorescent (EGFP<sup>high</sup>) subset includes mature and differentiated thrombocytes (<xref rid="bib21" ref-type="bibr">Ma et al., 2011</xref>).</p><p>Using flow cytometry, we identified EGFP<sup>low</sup> and EGFP<sup>high</sup> cells and sorted 188 cells from each population from a single kidney from a <italic>Tg(cd41:EGFP)</italic> reporter fish (<xref rid="fig1" ref-type="fig">Figure 1</xref>A; <xref rid="mmc1" ref-type="supplementary-material">Figures S1</xref>A–S1I). Each EGFP<sup>+</sup> cell was collected in a single well of a 96-well plate, and for each cell, its size (FSC), granularity (SSC), and EGFP fluorescence level were recorded. Single-cell mRNA-seq libraries were constructed and sequenced to a depth of around 2.5 million reads per library. Of 376 cells, 13 cells failed our quality control (QC) and were removed from further analysis (<xref rid="sec4" ref-type="sec">Experimental Procedures</xref>; <xref rid="mmc1" ref-type="supplementary-material">Figures S2</xref>A and S2B). For the remaining 363 cells, we accurately quantified between 1,000 and 6,000 genes per cell.</p></sec><sec id="sec2.2"><title>Ordering Hematopoietic Cells from a Single Kidney across Lineage Development</title><p>To identify groups of cells and order them in terms of their developmental progression, we used a multi-step approach. First, we used independent component analysis (ICA) to identify distinct factors that describe the variability of EGFP cells. ICA revealed four latent factors (hidden variables) that explain (1) a progression among EGFP<sup>low</sup> cells (“within_small_component”), (2) a switch from EGFP<sup>low</sup> cells toward EGFP<sup>high</sup> cells (“difference_component”), and (3) progression among the EGFP<sup>high</sup> cells (“within_large_component”). Finally, the fourth factor identified three outlier cells (“outlier_component”) (<xref rid="mmc1" ref-type="supplementary-material">Figure S3</xref>A).</p><p>To facilitate data depiction, we used non-linear dimensionality reduction (t-distributed stochastic neighbor embedding [t-SNE]; <xref rid="bib44" ref-type="bibr">Van der Maaten and Hinton, 2008</xref>) to represent the four latent factors in two dimensions (<xref rid="fig1" ref-type="fig">Figure 1</xref>B). ICA revealed a clear distinction between EGFP<sup>low</sup> and EGFP<sup>high</sup> cells, implying sharp divergence at the transcriptional level (<xref rid="mmc1" ref-type="supplementary-material">Figure S3</xref>A; <xref rid="fig1" ref-type="fig">Figure 1</xref>B).</p><p>In addition, EGFP<sup>low</sup> cells are a more heterogeneous group compared to EGFP<sup>high</sup> cells. To explore this further, we used hierarchical clustering to partition EGFP cells based on their independent components (<xref rid="mmc1" ref-type="supplementary-material">Figure S3</xref>B). Interestingly, whereas EGFP<sup>low</sup> cells were split into four distinct clusters (here named 1a, 1b, 2, and 3), EGFP<sup>high</sup> cells were all grouped into a single cluster (here named 4), confirming the substantial heterogeneity of the EGFP<sup>low</sup> population of cells (<xref rid="fig1" ref-type="fig">Figure 1</xref>C).</p><p>Differentiation of hematopoietic cells involves the acquisition of specific phenotypes that depend on the repression of genes characteristic of a multipotent cell state and expression of lineage-restricted genes (<xref rid="bib38" ref-type="bibr">Seita and Weissman, 2010</xref>). Thus, the whole process can be conceptualized as a temporal ordering of a highly coordinated transcriptional program through which each cell progresses. To examine the transcriptional transitions undergone by cd41-EGFP cells during differentiation, we ordered cells based on the cluster they belonged to, the latent factor that explains the variability of the cells within the cluster, and the level of EGFP fluorescence (details provided in the <xref rid="sec4" ref-type="sec">Experimental Procedures</xref>). Our model assumes gradual changes in gene expression during developmental progression of thrombocytes along a one-dimensional (i.e., non-branching) path. (We could not detect any apparent branch point in the data.) This ranking of cells through the entire process was treated as “pseudotime.”</p><p>To ensure our pseudotime ordering was stable, we also ordered the cells using an alternative method, a Bayesian Gaussian process latent variable model (<xref rid="bib43" ref-type="bibr">Titsias and Lawrence, 2010</xref>; see <xref rid="sec4" ref-type="sec">Experimental Procedures</xref>). Comparing the paths these orderings take when regressed into the t-SNE depiction, one can appreciate the similarity between them (<xref rid="fig2" ref-type="fig">Figure 2</xref>A). The two pseudotime orderings agreed very strongly (Spearman correlation 0.97; <xref rid="fig2" ref-type="fig">Figure 2</xref>B), giving us confidence in our method.</p><p>When presented in pseudotime, the expression of endogenous <italic>cd41</italic> (also known as <italic>itga2b</italic>) and <italic>EGFP</italic>, as well as EGFP fluorescence, recorded during sorting, were highly correlated and showed an expected increase through pseudotime (Spearman rho 0.85, 0.80, and 0.82, respectively) (<xref rid="fig2" ref-type="fig">Figure 2</xref>C). This supports our pseudotime ordering of the cells from the HSC to the differentiated thrombocyte extracted from a single kidney.</p></sec><sec id="sec2.3"><title>Inferring Cell States in the Myeloid Lineage</title><p>To define the identity of cell types within the five clusters, we evaluated the expression of orthologs of transcription factors and other genes known to be relevant in mammalian hematopoiesis, including the expression of early (cd61, also known as <italic>itgb3a/b</italic>) and late (cd42b, also known as <italic>gp1bb</italic>) markers of megakaryocyte differentiation (<xref rid="fig3" ref-type="fig">Figure 3</xref>). The panel of genes analyzed was representative of HSCs (<italic>Tal1</italic>, <italic>Lmo2</italic>, <italic>Lyl1</italic>, <italic>Gata2</italic>, <italic>Runx1</italic>, <italic>Meis1</italic>, <italic>C-myb</italic>, and <italic>Erg</italic>; <xref rid="bib2" ref-type="bibr">Capron et al., 2006</xref>, <xref rid="bib12" ref-type="bibr">Greig et al., 2008</xref>, <xref rid="bib20" ref-type="bibr">Loughran et al., 2008</xref>, <xref rid="bib25" ref-type="bibr">Orkin and Zon, 2008</xref>, <xref rid="bib31" ref-type="bibr">Pineault et al., 2002</xref>), megakaryocyte/erythroid (<italic>Fli1</italic>, <italic>Gfi1b</italic>, <italic>Gata1</italic>, <italic>Cd61</italic>, <italic>Cd42b</italic>, <italic>Vwf</italic>, and <italic>Selp</italic>; <xref rid="bib5" ref-type="bibr">Clay et al., 2001</xref>, <xref rid="bib25" ref-type="bibr">Orkin and Zon, 2008</xref>, <xref rid="bib32" ref-type="bibr">Poirault-Chassac et al., 2010</xref>, <xref rid="bib36" ref-type="bibr">Schick et al., 1993</xref>), and myeloid- (<italic>Gfi1</italic>, <italic>Pu.1</italic> also known as <italic>spi1a/b</italic>, and <italic>Cebp1</italic>; <xref rid="bib41" ref-type="bibr">Tenen et al., 1997</xref>, <xref rid="bib50" ref-type="bibr">Zeng et al., 2004</xref>) lineage-affiliated genes.</p><p>For each gene, we assessed the level of its expression in pseudotime, as well as the fraction of cells that expressed the gene of interest in each of the clusters (<xref rid="fig3" ref-type="fig">Figure 3</xref>). For example, <italic>c-myb</italic> was highly expressed in cluster 1a, as well as in clusters 1b, 2, and 3, but was downregulated in cluster 4. This is in line with previous reports that <italic>C-myb</italic> is expressed in immature hematopoietic cells and is downregulated during differentiation (<xref rid="bib12" ref-type="bibr">Greig et al., 2008</xref>). Cells in cluster 1a had relatively high expression of <italic>lmo2</italic>, <italic>tal1</italic>, and <italic>meis1</italic>. These genes, together with <italic>fli1</italic>, showed a similar distribution of expression across pseudotime, whereas <italic>gata2</italic> was more restricted to cluster 1a. The mammalian HSC genes <italic>runx1</italic> and <italic>erg</italic> were expressed at a relatively low level overall, and in a small fraction of cells within all clusters. Overall, most of the mammalian HSC marker genes examined are expressed in cluster 1a, and to a lesser degree in 1b, 2, and 3.</p><p>In contrast, <italic>Gata1</italic> and <italic>Gfi1b</italic> are known to be expressed at high levels in the erythroid and megakaryocyte lineages (<xref rid="bib25" ref-type="bibr">Orkin and Zon, 2008</xref>, <xref rid="bib45" ref-type="bibr">Vassen et al., 2007</xref>) but not in HSCs. In our dataset, <italic>gata1a</italic> and <italic>gfi1b</italic> were expressed in all clusters except cluster 1a. Furthermore, expression of both early (<italic>itgb3a/b</italic>) and late (<italic>gp1bb</italic>) markers of megakaryocyte differentiation started very early and peaked late in pseudotime, confirming that more mature thrombocytes are largely confined to cluster 4.</p><p>We also assessed the expression of two well-known platelet genes, <italic>vWf</italic> (von Willebrand factor) and <italic>selp</italic> (P-selectin), through pseudotime (<xref rid="fig3" ref-type="fig">Figure 3</xref>). Our analysis revealed that, contrary to previous reports (<xref rid="bib4" ref-type="bibr">Carrillo et al., 2010</xref>), thrombocytes in zebrafish do not express von Willebrand factor and P-selectin. This was confirmed by qPCR analysis of <italic>cd41</italic> EGFP<sup>high</sup> thrombocytes from zebrafish kidney. We found, however, that <italic>vWf</italic> was expressed in the whole kidney sample and in fli1:GFP positive cells sorted from <italic>Tg(fli:EGFP)</italic> fish, suggesting that the <italic>vWf</italic> expression pattern differs somewhat in zebrafish compared to mammals.</p><p>Surprisingly, myeloid lineage-affiliated genes (e.g., <italic>spi1</italic>, <italic>gfi1</italic>, and <italic>cebp1</italic>) were largely absent across all cells (<xref rid="fig3" ref-type="fig">Figure 3</xref>). This suggests that there is no common myeloid progenitor population in this dataset, which charts a continuous HSC to thrombocyte pathway. Altogether, our data are consistent with the notion that cells from cluster 1a belong to HSCs that transition directly to erythroid-thrombocyte progenitor cells, possibly circumventing the CMP step. Although this is surprising, there are other reports of direct, unconventional, HSC to megakaryocyte-erythroid progenitor transitions, such as a recent report in mouse (<xref rid="bib13" ref-type="bibr">Guo et al., 2013</xref>).</p><p>Identification of these progenitor and differentiated cell types prompted us to carry out additional analyses of the sets of genes that strongly correlate with the individual cell types. We used a machine learning method, random forest feature importance, to find genes whose expression “marks” distinct clusters of cells. The unique sets of genes expressed in each of the cell types provide an opportunity to reveal novel markers of the identified cell types, and at the same time, provide more insight into their biological function.</p><p>Among the numerous newly identified cell-type markers (<xref rid="mmc2" ref-type="supplementary-material">Table S1</xref>), we found several of particular interest (<xref rid="fig4" ref-type="fig">Figure 4</xref>A). Ccr9a is a member of the beta chemokine receptor family and is known to be expressed in HSCs (<xref rid="bib49" ref-type="bibr">Wright et al., 2002</xref>); our data show that <italic>ccr9a</italic> expression is highly correlated with cluster 1a (<xref rid="fig4" ref-type="fig">Figure 4</xref>B). Transcription elongation factor A (SII), <italic>tcea3</italic>, was specifically expressed in cluster 1b (<xref rid="fig4" ref-type="fig">Figure 4</xref>B). Cells from cluster 1b can also be sorted by combining expression of plasminogen receptor gene (<italic>plgrkt</italic>) and <italic>ascc1</italic> (<xref rid="fig4" ref-type="fig">Figure 4</xref>B). Good marker genes for cluster 2 included <italic>e2f8</italic>, which encodes a protein involved in progression through the cell cycle (<xref rid="bib8" ref-type="bibr">Deng et al., 2010</xref>) and <italic>top2a</italic>, a DNA topoisomerase involved in processes such as chromosome condensation and chromatid separation (<xref rid="bib10" ref-type="bibr">Downes et al., 1994</xref>) (<xref rid="fig4" ref-type="fig">Figure 4</xref>B). Interestingly, the overrepresented gene ontology (GO) enrichment terms for cluster 2 included cell division and cell cycle (<xref rid="fig4" ref-type="fig">Figure 4</xref>A), suggesting that an expansion phase precedes lineage commitment and terminal differentiation of thrombocytes.</p><p>To experimentally validate the prediction of greater proliferation in this progenitor population, we sorted cells from clusters 1a/1b/2 versus 3 and 4, by distinguishing these three populations based on EGFP fluorescence, and SSC and FSC (<xref rid="mmc1" ref-type="supplementary-material">Figures S4</xref>A–S4G). We compared the cell cycle distributions of the sorted populations using propidium iodide (PI) staining. The combined cells from clusters 1a/1b/2 had a significantly higher proportion of cells in S and G2/M phase compared to clusters 3 and 4 (<xref rid="fig4" ref-type="fig">Figure 4</xref>C), validating our finding that these cells proliferate faster.</p><p>These results show that expression of EGFP together with SSC and FSC values could be used to efficiently separate cells from clusters 3 and 4 from the early progenitor populations (1a/1b/2) in the cd41 reporter line (<xref rid="mmc1" ref-type="supplementary-material">Figures S4</xref> and <xref rid="mmc1" ref-type="supplementary-material">S5</xref>). Additional markers for cluster 3 included combined high expression of <italic>fzd8b</italic> and no expression of <italic>mibp</italic> (<xref rid="fig4" ref-type="fig">Figure 4</xref>B). For cluster 4, a high level of cd41 uniquely marks this population.</p><p>Finally, we also assessed a unique set of genes expressed by the three outlier cells. GO enrichment analysis of their marker genes yielded only three statistically significant GO terms, all linked with immunity (<xref rid="fig4" ref-type="fig">Figure 4</xref>A). One plausible explanation is that these outlier cells represent macrophages that have engulfed or are attached to thrombocytes and hence retained a high level of EGFP fluorescence. Indeed, the outlier cells expressed an array of macrophage/monocyte affiliated genes such as <italic>mpeg</italic> (macrophage expressed gene 1), <italic>csf1r</italic> (colony-stimulating factor 1 receptor), <italic>csf3r</italic> (colony-stimulating factor 3 receptor) etc. Furthermore, compared to all other cells, the outlier cells had remarkably high FSC and SSC values, characteristic of macrophages (<xref rid="fig4" ref-type="fig">Figure 4</xref>D).</p></sec><sec id="sec2.4"><title>Validation of Developmental Progression from the Kidney and Circulation</title><p>Importantly, we validated many of our findings in a second set of single cell transcriptomics experiments on kidney cells, as well as circulating cells, from another fish. We sorted an additional 92 cells from cluster 1a/1b/2 (named here EarlyEnriched), 46 EGFP<sup>low</sup> cells and 46 EFP<sup>high</sup> cells from the kidney of another <italic>Tg(cd41:EGFP)</italic> fish. We also sorted 24 EGFP<sup>low</sup> and 68 EGFP<sup>high</sup> circulating cells from the same fish (<xref rid="mmc1" ref-type="supplementary-material">Figure S6</xref>A). Our analysis confirmed that the pattern of ICA follows the same structure as observed in the previous experiment (<xref rid="mmc1" ref-type="supplementary-material">Figure S6</xref>B). This means that the cell populations and their relative relationships are conserved in this biological replicate. Similarly, the pseudotime ordering of EarlyEnriched, EGFP<sup>low</sup>, and EGFP<sup>high</sup> cells in the kidney recapitulated patterns we identified in the initial experiment (<xref rid="fig5" ref-type="fig">Figures 5</xref>A–5F).</p><p>In addition, we discovered that EGFP<sup>high</sup> cells in circulation are transcriptionally identical to EGFP<sup>high</sup> cells in the kidney, with no significant change in the number of expressed genes (<xref rid="fig5" ref-type="fig">Figure 5</xref>B), RNA content (<xref rid="fig5" ref-type="fig">Figure 5</xref>C), or any gene’s expression pattern (likelihood ratio test, corrected for multiple testing with Holm-Sidak). We concluded, therefore, that the thrombocytes exit the kidney in a fully mature state and are maintained in a transcriptionally active state in circulation.</p><p>In both datasets, the total number of genes and total mRNA content expressed per cell were correlated with its differentiation state (<xref rid="mmc1" ref-type="supplementary-material">Figure S7</xref>). This was not due to a difference in the sequencing depth or cell size (<xref rid="mmc1" ref-type="supplementary-material">Figure S7</xref>). Instead, it represents a biological difference between cells during development. This supports the idea that more differentiated, post-mitotic cells (clusters 3 and 4) have a specialized transcriptional program with expression of a small, focused set of genes (<xref rid="mmc1" ref-type="supplementary-material">Figure S7</xref>).</p></sec><sec id="sec2.5"><title>Transcriptional Modules Related to Growth and Proliferation in the Thrombocyte Developmental Gene Expression Program</title><p>To find genes with similar trends in expression across pseudotime, we used a mixtures of hierarchical Gaussian processes model to cluster the pseudotime series (<xref rid="bib14" ref-type="bibr">Hensman et al., 2015</xref>). We identified 130 genes that are dynamically expressed through pseudotime. Clustering of these genes revealed three distinct patterns of their progression during differentiation (<xref rid="fig6" ref-type="fig">Figure 6</xref>A; <xref rid="mmc3" ref-type="supplementary-material">Table S2</xref>). Genes upregulated early in pseudotime and then downregulated later (group I) were significantly enriched with the GO term “nucleic acid binding” and “chromosome maintenance” (<xref rid="fig6" ref-type="fig">Figures 6</xref>B and 6C; <xref rid="mmc3" ref-type="supplementary-material">Table S2</xref>), possibly reflecting the increased proliferation of cells earlier in pseudotime. Genes gradually downregulated through pseudotime (group II) were highly enriched with the GO terms “eukaryotic translation elongation,” “ribosomes” etc. (<xref rid="fig6" ref-type="fig">Figures 6</xref>B and 6C; <xref rid="mmc3" ref-type="supplementary-material">Table S2</xref>). Expression of these genes was highly correlated with the general trend of decreased RNA content over pseudotime (Spearman rho = 0.85), therefore suggesting a regulatory loop between the total RNA content in the cell and expression of genes that encode proteins relevant for ribosome synthesis. Finally, genes upregulated early and then maintained at a high level (group III) were highly enriched with the GO terms “ECM-receptor interaction,” “platelet aggregation,” and “hemostasis,” pointing to the genes important for thrombocyte function (<xref rid="fig6" ref-type="fig">Figures 6</xref>B and 6C; <xref rid="mmc3" ref-type="supplementary-material">Table S2</xref>). Taken together, our analysis suggests that differentiation of thrombocytes is governed by coordinated transcriptional programs that limit the proliferation of cells and their translational capacity while simultaneously promoting genes relevant for thrombocyte function.</p></sec><sec id="sec2.6"><title>Single Cell Gene Expression Patterns of Whole-Genome Duplicated Genes</title><p>Gene duplication is a common event in eukaryotic genomes (<xref rid="bib22" ref-type="bibr">Meyer and Schartl, 1999</xref>) and due to the teleost-specific genome duplication around 26% (i.e., 3,440) (<xref rid="bib15" ref-type="bibr">Howe et al., 2013</xref>) of zebrafish genes are duplicated. Gene duplicates that originate from genome duplication are called ohnologs. To assess the use of duplicated genes during thrombopoiesis in zebrafish, we examined the expression of ohnologs in each of the 363 <italic>cd41</italic>:EGFP cells. Of ∼8,000 ohnolog categories (<xref rid="bib15" ref-type="bibr">Howe et al., 2013</xref>), we looked at 3,034 ohnolog categories that have only been duplicated once (ohnolog gene pairs). Of these 3,034 ohnologs (<xref rid="bib15" ref-type="bibr">Howe et al., 2013</xref>), 2,107 were not expressed in our dataset. However the remaining 927 pairs can be divided into the following three major groups: (1) expression of ohnologs is mutually exclusive in individual cells (n = 177) (<xref rid="fig7" ref-type="fig">Figures 7</xref>A and 7B; <xref rid="mmc4" ref-type="supplementary-material">Table S3</xref>). In this group, the expression of any one ohnolog appeared to be an independent event with an equal probability of happening. This suggests selective activation or silencing of these ohnologs in individual cells; (2) only one of the ohnologs is expressed in all cells (n = 430), <xref rid="fig7" ref-type="fig">Figures 7</xref>A and 7B; <xref rid="mmc4" ref-type="supplementary-material">Table S3</xref>), and (3) both ohnologs are equally expressed in all cells (n = 218), (<xref rid="fig7" ref-type="fig">Figures 7</xref>A and 7B; <xref rid="mmc4" ref-type="supplementary-material">Table S3</xref>). No patterns of ohnolog use over pseudotime were observed.</p></sec></sec><sec id="sec3"><title>Discussion</title><p>Here we show the power of single cell transcriptome analysis to decipher the kinetics of hematopoietic lineage development. We ordered cd41 cells by their progression through differentiation based on gene expression profiles. Our analysis illustrates the continual nature of this process, where cells progressively transit through five transcriptional states that result in the generation of mature thrombocytes.</p><p>Interestingly, myeloid lineage-affiliated genes were largely absent across all cells, suggesting direct HSC to thrombocyte-erythroid progenitor transition. The model of hematopoiesis generated recently, using single cells from over ten hematopoietic populations in mouse, implies that the megakaryocyte-erythroid lineage is closely linked to long-term repopulating HSCs and separates early from the lympho-myeloid lineage (<xref rid="bib13" ref-type="bibr">Guo et al., 2013</xref>). The identification of platelet-primed stem cells within vWf-expressing long-term HSCs further confirmed that commitment to the megakaryocyte lineage starts in the most primitive stem cell compartment (<xref rid="bib35" ref-type="bibr">Sanjuan-Pla et al., 2013</xref>). Although in our dataset vWf was not expressed in any of the identified cell populations, the low expression of some of the thrombocyte lineage-affiliated genes in cluster 1a suggests that using our sorting strategy we are possibly capturing thrombocyte-primed stem cells. Therefore, HSCs in cluster 1a may represent a biased subpopulation within the wider pool of hematopoietic stem/progenitor cells present in the zebrafish kidney. Nevertheless, the gradual transition of cells during thrombocyte lineage development that we see in our dataset (e.g., gradual changes in the total number of genes as well as the total mRNA content) suggest that we do capture a continuous spectra of cells and that the common myeloid stage is not an obligatory step during thrombopoiesis.</p><p>We also show that although each of the identified transcriptional states was characterized by substantial heterogeneity in the expression of the key lineage regulators, the underlying transcriptional program was highly coordinated. It included the simultaneous increase in the expression of genes important for thrombocyte function and suppression of genes relevant in cell proliferation and ribosomal biogenesis. Interestingly, although the maturation of thrombocytes was completed in the kidney, they maintained a transcriptionally active state in circulation. We did not, however, detect any qualitative or quantitative difference in the gene expression between circulating and kidney-based EGFP<sup>high</sup> thrombocytes. Surprisingly, unlike mammalian platelets, which have abundant expression of vWf, thrombocytes in zebrafish do not express vWf. Instead, our analysis suggests that other cells within the kidney marrow, such as endothelial cells (<italic>fli1:GFP</italic> positive cells), express vWf in zebrafish. Finally, we assessed use of duplicated genes during thrombopoiesis in zebrafish and identified patterns of their expression that would not be possible using a bulk transcriptomics approach.</p><p>We used single-cell RNA-seq of zebrafish kidney cells to resolve the cellular hierarchy of lineage development in the myeloid branch of hematopoiesis and propose a refined model of developmental progression of hematopoietic cells.</p><p>Our study addresses some of the basic questions of regulation of differentiation, both at the molecular and cellular levels. In this study, we focused on zebrafish thrombocyte development; however, a similar approach could be used in other systems and cell types.</p></sec><sec id="sec4"><title>Experimental Procedures</title><sec id="sec4.1"><title>Zebrafish Strains and Maintenance</title><p>The maintenance of wild-type (Tubingen Long Fin) and transgenic zebrafish Tg(<italic>cd41:GFP</italic>) lines were performed in accordance with EU regulations on laboratory animals, as previously described (<xref rid="bib1" ref-type="bibr">Bielczyk-Maczyńska et al., 2014</xref>).</p></sec><sec id="sec4.2"><title>Single-Cell Sorting and Whole Transcriptome Amplification</title><p>A single kidney from heterozygote <italic>Tg</italic>(<italic>cd41:EGFP</italic>) or wild-type fish was dissected and carefully passed through a strainer using the plunger of a 1 ml syringe. In the follow-up experiment, circulating GFP-positive cells were collected from the dissected heart of the same fish. Cells were collected in cold 1× PBS/5% fetal bovine serum. The kidney of a non-transgenic line was used to set up the gating and exclude autofluorescent cells. Dead cells were excluded based on PI staining. Individual cells were sorted using a Becton Dickinson Influx sorter with 488- and 561-nm lasers (<xref rid="bib37" ref-type="bibr">Schulte et al., 2015</xref>) and collected in a single well of a 96-well plate containing 2.3 μl of 0.2% Triton X-100 supplemented with 1 U/μl SUPERase In RNase inhibitor (Ambion). At the same time, information about cell size and granularity and the level of the fluorescence were recorded. Whole transcriptome amplification and library preparation was performed using the Smart-seq2 protocol (<xref rid="bib30" ref-type="bibr">Picelli et al., 2014</xref>, <xref rid="bib29" ref-type="bibr">Picelli et al., 2013</xref>), with ERCC spike-in controls added at the same time as the oligo-dT and dNTP mixture. Twenty-five PCR cycles were performed during the amplification.</p></sec><sec id="sec4.3"><title>Cell Cycle Analysis</title><p>GFP-positive cells from <italic>Tg</italic>(<italic>cd41:EGFP</italic>) kidney suspension were sorted using a Mo-Flo XDP (Beckman Coulter) with 488-, 561-, and 640-nm lasers. Cells were centrifuged at 1,200 rpm for 10 min at 4°C, resuspended in 100 μl 1× PBS and fixed by adding 300 μl ethanol. Cells were fixed overnight at 4°C, washed twice in 1× PBS, and re-suspended in 500 μl PI solution (25 μg/ml PI, 0.1% Triton X-100, 0.1% sodium citrate). Cells were incubated for 3 hr with RNase A (Sigma) and analyzed by BD LSR Fortessa (Becton Dickinson). Data were analyzed using FlowJo software.</p></sec><sec id="sec4.4"><title>Cytology</title><p>Sorted EGFP-positive cells were concentrated by cytocentrifugation at 350 rpm for 5 min onto SuperFrostPlus slides using a Shandon Cytospin 3 cytocentrifuge. Slides were fixed for 3 min in methanol and stained with May-Grünwald Giemsa (Sigma) as described elsewhere (<xref rid="bib40" ref-type="bibr">Stachura et al., 2009</xref>). Images were captured as described elsewhere (<xref rid="bib1" ref-type="bibr">Bielczyk-Maczyńska et al., 2014</xref>).</p></sec><sec id="sec4.5"><title>Verification of RNA-Seq Data with qPCR</title><p>GFP-positive cells from <italic>Tg</italic>(<italic>cd41:EGFP</italic>) and <italic>Tg</italic>(<italic>fli1:EGFP</italic>) kidney suspensions were sorted using a Mo-Flo XDP (Beckman Coulter), along with an equal number of viable cells from the whole kidney, into 75 μl RLT buffer (QIAGEN) containing 1% β-mercaptoethanol. mRNA was extracted using Oligo (dT)<sub>25</sub> Dynabeads (Ambion) and cDNA was prepared using SuperScript VILO (Invitrogen), according to the manufacturers’ instructions. qPCR reactions were performed using the 7900HT Real Time system (Life Technologies) with primers for <italic>vWf</italic> (F: CGGCAGCACATACACACATT and R: CGTTCCATCCACAGAGAGGT) and two housekeeping genes (eif1a F: GAGAAGTTCGAGAAGGAAGC and R: CGTAGTATTTGCTGGTCTCG, and b-actin F: CGAGCAGGAGATGGGAACC and R: CAACGGAAACGCTCATTGC). The ΔΔCt method was used for data analysis.</p></sec><sec id="sec4.6"><title>Single-Cell RNA-Seq Data Processing</title><p>Reads from RNA-seq were aligned to the zebrafish genome (Zv9.77) combined with sequences for eGFP and ERCC spike-ins as artificial chromosomes, using STAR (version 2.3; (<xref rid="bib9" ref-type="bibr">Dobin et al., 2013</xref>). The Ensembl Genes annotation track from UCSC was used with the read_distribution.py tool from the RSeQC tool suite (<xref rid="bib47" ref-type="bibr">Wang et al., 2012</xref>) to generate quality control information. Gene expression was quantified using the Salmon (<xref rid="bib27" ref-type="bibr">Patro et al., 2015</xref>) reads mode of Sailfish (<xref rid="bib26" ref-type="bibr">Patro et al., 2014</xref>; parameter -l IU) using Zv9 cDNA sequences from Ensembl version 77 as transcript sequences, together with ERCC spike-in and eGFP sequences as artificial transcripts. Based on comparison with empty control wells, samples with less than 50,000 paired reads and 1,000 expressed genes were considered unfit and were excluded from further analysis (<xref rid="mmc1" ref-type="supplementary-material">Figure S2</xref>).</p><p>For the follow-up experiment, expression was quantified the same way. We used a different stock and concentration of ERCC spike-ins, which changed the scales of the QC values. For these samples, we excluded cells with less than 200,000 paired reads and less than 150 expressed genes (<xref rid="mmc1" ref-type="supplementary-material">Figure S6</xref>).</p><p>Downstream analysis was performed using Transcripts per million (TPM) values reported by Salmon. The TPM unit is a measure of relative abundance of a gene, which is stable across samples (<xref rid="bib19" ref-type="bibr">Li and Dewey, 2011</xref>, <xref rid="bib46" ref-type="bibr">Wagner et al., 2012</xref>). Before analysis expression for endogenous spike-ins were filtered out for each cell, and the TPM for each cell was rescaled to sum to a million. This gives us the interpretation that TPM of a gene will correspond to the concentration of mRNAs from a gene in a given cell.</p><p>Unless stated otherwise, for all analyses, we filtered out genes expressed at a level higher than 1 TPM in only less than three cells, which leaves 20,556 genes.</p></sec><sec id="sec4.7"><title>Identifying Processes and Ordering Cells by Hidden Factors</title><p>We used ICA (<xref rid="bib17" ref-type="bibr">Hyvärinen and Oja, 2000</xref>) to identify four latent factors (hidden variables modeling the data), as implemented in scikit-learn (with parameter random_state = 3,984 for the sake of reproducibility). The choice of four components was based on testing between one and ten components, and seeing diminishing returns on the Frobenius norm reconstruction error past four components. One latent factor explains a progression among EGFP<sup>low</sup> cells; another factor explains a switch from EGFP<sup>low</sup> cells toward the population of EGFP<sup>high</sup> cells. A third factor explains progression among EGFP<sup>high</sup> cells. The fourth factor identifies three outlier cells. We used the fluorescence levels of GFP to flip the orientation of the latent factors so that a higher factor value always corresponded to a higher GFP value. Because these factors are orthogonal, they are statistically independent. In other words, there are three distinct processes happening sequentially. We performed hierarchical Ward clustering (<xref rid="bib48" ref-type="bibr">Ward, 1963</xref>) of the cells in the four-dimensional ICA space, and assigned the cells to six clusters. (For exact commands, see Notebook 1 in <xref rid="mmc6" ref-type="supplementary-material">Data S2</xref>.) Based on which cluster the cells belonged to, and which factor explains the variability of the cells of that cluster, we ordered cells along this three-stage progression. This ranking of cells through the entire process was treated as pseudotime. (For exact commands, see Notebook 3 in <xref rid="mmc6" ref-type="supplementary-material">Data S2</xref>.)</p><p>As an alternative way to estimate a pseudotime, we applied a Bayesian Gaussian process latent variable model with a one-dimensional latent variable (<xref rid="bib43" ref-type="bibr">Titsias and Lawrence, 2010</xref>). Briefly, the Bayesian GPLVM will infer a nonlinear function from an unobserved latent space to an observed high-dimensional space, using inducing inputs that are variationally inferred, which helps smooth the data and speed up computation. In our case, the latent space is the one-dimensional pseudotime, and the non-linear function will be a mapping from pseudotime to gene expression values. We used the BayesianGPLVM implementation in the GPy package (<xref rid="bib42" ref-type="bibr">The GPy authors, n.d.</xref>) using a Radial Basis Function (RBF) kernel on the log-transformed TPM values, all other parameters default. Without any information about the EGFP expression, the BayesianGPLVM recovers our original ordering, up to orientation (Spearman correlation 0.97; <xref rid="fig2" ref-type="fig">Figure 2</xref>B) (Notebook 7 in <xref rid="mmc6" ref-type="supplementary-material">Data S2</xref>).</p><p>To depict the structure of the data in a friendly way, we performed t-distributed stochastic neighbor embedding (t-SNE) (<xref rid="bib44" ref-type="bibr">Van der Maaten and Hinton, 2008</xref>) of the four latent factors into two dimensions. The goal of the t-SNE algorithm is to attempt to preserve both global and local structures of higher dimensional data in two dimensions. It additionally tries to not crowd areas with too many points, making them hard to see. We set the perplexity parameter to 75 and used a fixed random seed to make sure the t-SNE plot would be reproducible (parameter random_state = 254 in the scikit-learn implementation of t-SNE).</p><p>We can depict the inferred pseudotime by regressing it into the two-dimensional tSNE space (<xref rid="fig2" ref-type="fig">Figure 2</xref>A) and can see how well the two methods of constructing pseudotime agrees.</p></sec><sec id="sec4.8"><title>Marker Gene Discovery</title><p>To discover marker genes for the clusters of cells, we trained a random forest model for each cluster versus the rest of the cells. We used the Gini feature importance scores for each gene to order genes by how well they can distinguish a cluster from the rest of the cells. We used the ExtraTreesClassifier (<xref rid="bib11" ref-type="bibr">Geurts et al., 2006</xref>) implementation in the Python machine learning package scikit-learn (<xref rid="bib28" ref-type="bibr">Pedregosa et al., 2011</xref>), with the parameter n_estimators = 100,000. (For the exact commands, see Notebook 2 in <xref rid="mmc6" ref-type="supplementary-material">Data S2</xref>.)</p></sec><sec id="sec4.9"><title>Pseudotime Analysis</title><p>We treated the pseudotime progression order of the cells as a time series, and for each gene trained two Gaussian processes (GPs): one with a radial basis function (RBF) kernel (which can model change over time) and one with a constant kernel (which assumes that the expression of a gene does not change over time). After optimizing parameters for both models, we filtered the genes by the ratios of the likelihoods of the models. If the RBF kernel GP has a higher likelihood than the constant kernel GP, we can conclude that the gene in question has expression that is dynamic in time. Once we had identified genes that were pseudotime-dependent, we applied the mixtures of hierarchical Gaussian processes model to identify groupings of genes with similar pseudotime expression patterns (<xref rid="bib14" ref-type="bibr">Hensman et al., 2015</xref>). All functional enrichment analysis was performed with the gProfiler (<xref rid="bib33" ref-type="bibr">Reimand et al., 2011</xref>) web service with the standard gene list as background (see Notebook 4 in <xref rid="mmc6" ref-type="supplementary-material">Data S2</xref> for exact commands).</p></sec><sec id="sec4.10"><title>Classification of Ohnolog Gene Pairs</title><p>We obtained the list of duplicated genes arising from the teleost-specific genome duplication event from (<xref rid="bib15" ref-type="bibr">Howe et al., 2013</xref>). We filtered the list to only retain pairs of genes whose IDs were present in version 77 of Ensembl. For these genes, we binarized the expression to “expressed” or “not expressed” in each cell based on whether the TPM was greater than 1. Using these binary values, for each Ohnolog pair we counted cells expressing either member of the pair, both members of the pair, or none of the members in the pair. Ohnolog pairs in which none of the members were expressed in more than 300 cells and were annotated as “Not expressed.” We defined a value “both_min_diff” as the difference between the smallest number of cells expressing only one of the members in a pair, and the number of cells expressing both members of the pair. Ohnolog pairs with a “both_min_diff”-value larger than 15 were annotated as “XOR Ohnologs.” To identify Ohnolog pairs in which only one member was used, we looked at the difference between the largest number of cells using one member compared to the largest number of cells using the other member. If this difference was larger than 60 cells, the Ohnolog pair was considered a “Single Ohnolog.” The remaining cells were dubbed “Mixed Ohnologs,” meaning cells with a mixture of both members of a pair. (See Notebook 5 in <xref rid="mmc6" ref-type="supplementary-material">Data S2</xref> for exact commands.)</p><p>All analysis scripts are provided as IPython notebooks in the supplemental information (<xref rid="mmc5" ref-type="supplementary-material">Data S1</xref>, Sample Information) together with a table of detailed information of each sample (<xref rid="mmc6" ref-type="supplementary-material">Data S2</xref>, Analysis Files).</p></sec></sec><sec id="sec5"><title>Author Contributions</title><p>I.C.M, C.L., and L.F. performed experiments. V.S. carried out the analysis. I.C.M, C.L., V.S., F.H., S.T., and A.C. contributed to the discussion of the results. T.V. oversaw implementation of the scRNA-seq pipeline. I.C.M, C.L., V.S., and A.C. wrote the manuscript. S.H. edited the manuscript. A.C. conceived the study. All authors approved the final version of the manuscript.</p></sec></body><back><ref-list><title>References</title><ref id="bib1"><element-citation publication-type="journal" id="sref1"><person-group person-group-type="author"><name><surname>Bielczyk-Maczyńska</surname><given-names>E.</given-names></name><name><surname>Serbanovic-Canic</surname><given-names>J.</given-names></name><name><surname>Ferreira</surname><given-names>L.</given-names></name><name><surname>Soranzo</surname><given-names>N.</given-names></name><name><surname>Stemple</surname><given-names>D.L.</given-names></name><name><surname>Ouwehand</surname><given-names>W.H.</given-names></name><name><surname>Cvejic</surname><given-names>A.</given-names></name></person-group><article-title>A loss of function screen of identified genome-wide association study Loci reveals new genes controlling hematopoiesis</article-title><source>PLoS Genet.</source><volume>10</volume><year>2014</year><fpage>e1004450</fpage><pub-id pub-id-type="pmid">25010335</pub-id></element-citation></ref><ref id="bib2"><element-citation publication-type="journal" id="sref2"><person-group person-group-type="author"><name><surname>Capron</surname><given-names>C.</given-names></name><name><surname>Lécluse</surname><given-names>Y.</given-names></name><name><surname>Kaushik</surname><given-names>A.L.</given-names></name><name><surname>Foudi</surname><given-names>A.</given-names></name><name><surname>Lacout</surname><given-names>C.</given-names></name><name><surname>Sekkai</surname><given-names>D.</given-names></name><name><surname>Godin</surname><given-names>I.</given-names></name><name><surname>Albagli</surname><given-names>O.</given-names></name><name><surname>Poullion</surname><given-names>I.</given-names></name><name><surname>Svinartchouk</surname><given-names>F.</given-names></name></person-group><article-title>The SCL relative LYL-1 is required for fetal and adult hematopoietic stem cell function and B-cell differentiation</article-title><source>Blood</source><volume>107</volume><year>2006</year><fpage>4678</fpage><lpage>4686</lpage><pub-id pub-id-type="pmid">16514064</pub-id></element-citation></ref><ref id="bib3"><element-citation publication-type="journal" id="sref3"><person-group person-group-type="author"><name><surname>Carradice</surname><given-names>D.</given-names></name><name><surname>Lieschke</surname><given-names>G.J.</given-names></name></person-group><article-title>Zebrafish in hematology: sushi or science?</article-title><source>Blood</source><volume>111</volume><year>2008</year><fpage>3331</fpage><lpage>3342</lpage><pub-id pub-id-type="pmid">18182572</pub-id></element-citation></ref><ref id="bib4"><element-citation publication-type="journal" id="sref4"><person-group person-group-type="author"><name><surname>Carrillo</surname><given-names>M.</given-names></name><name><surname>Kim</surname><given-names>S.</given-names></name><name><surname>Rajpurohit</surname><given-names>S.K.</given-names></name><name><surname>Kulkarni</surname><given-names>V.</given-names></name><name><surname>Jagadeeswaran</surname><given-names>P.</given-names></name></person-group><article-title>Zebrafish von Willebrand factor</article-title><source>Blood Cells Mol. Dis.</source><volume>45</volume><year>2010</year><fpage>326</fpage><lpage>333</lpage><pub-id pub-id-type="pmid">21035359</pub-id></element-citation></ref><ref id="bib5"><element-citation publication-type="journal" id="sref5"><person-group person-group-type="author"><name><surname>Clay</surname><given-names>D.</given-names></name><name><surname>Rubinstein</surname><given-names>E.</given-names></name><name><surname>Mishal</surname><given-names>Z.</given-names></name><name><surname>Anjo</surname><given-names>A.</given-names></name><name><surname>Prenant</surname><given-names>M.</given-names></name><name><surname>Jasmin</surname><given-names>C.</given-names></name><name><surname>Boucheix</surname><given-names>C.</given-names></name><name><surname>Le Bousse-Kerdilès</surname><given-names>M.C.</given-names></name></person-group><article-title>CD9 and megakaryocyte differentiation</article-title><source>Blood</source><volume>97</volume><year>2001</year><fpage>1982</fpage><lpage>1989</lpage><pub-id pub-id-type="pmid">11264162</pub-id></element-citation></ref><ref id="bib6"><element-citation publication-type="journal" id="sref6"><person-group person-group-type="author"><name><surname>Cvejic</surname><given-names>A.</given-names></name></person-group><article-title>Mechanisms of fate decision and lineage commitment during haematopoiesis</article-title><source>Immunol. Cell Biol.</source><year>2015</year></element-citation></ref><ref id="bib7"><element-citation publication-type="journal" id="sref7"><person-group person-group-type="author"><name><surname>Debili</surname><given-names>N.</given-names></name><name><surname>Robin</surname><given-names>C.</given-names></name><name><surname>Schiavon</surname><given-names>V.</given-names></name><name><surname>Letestu</surname><given-names>R.</given-names></name><name><surname>Pflumio</surname><given-names>F.</given-names></name><name><surname>Mitjavila-Garcia</surname><given-names>M.T.</given-names></name><name><surname>Coulombel</surname><given-names>L.</given-names></name><name><surname>Vainchenker</surname><given-names>W.</given-names></name></person-group><article-title>Different expression of CD41 on human lymphoid and myeloid progenitors from adults and neonates</article-title><source>Blood</source><volume>97</volume><year>2001</year><fpage>2023</fpage><lpage>2030</lpage><pub-id pub-id-type="pmid">11264167</pub-id></element-citation></ref><ref id="bib8"><element-citation publication-type="journal" id="sref8"><person-group person-group-type="author"><name><surname>Deng</surname><given-names>Q.</given-names></name><name><surname>Wang</surname><given-names>Q.</given-names></name><name><surname>Zong</surname><given-names>W.-Y.</given-names></name><name><surname>Zheng</surname><given-names>D.-L.</given-names></name><name><surname>Wen</surname><given-names>Y.-X.</given-names></name><name><surname>Wang</surname><given-names>K.-S.</given-names></name><name><surname>Teng</surname><given-names>X.-M.</given-names></name><name><surname>Zhang</surname><given-names>X.</given-names></name><name><surname>Huang</surname><given-names>J.</given-names></name><name><surname>Han</surname><given-names>Z.-G.</given-names></name></person-group><article-title>E2F8 contributes to human hepatocellular carcinoma via regulating cell proliferation</article-title><source>Cancer Res.</source><volume>70</volume><year>2010</year><fpage>782</fpage><lpage>791</lpage><pub-id pub-id-type="pmid">20068156</pub-id></element-citation></ref><ref id="bib9"><element-citation publication-type="journal" id="sref9"><person-group person-group-type="author"><name><surname>Dobin</surname><given-names>A.</given-names></name><name><surname>Davis</surname><given-names>C.A.</given-names></name><name><surname>Schlesinger</surname><given-names>F.</given-names></name><name><surname>Drenkow</surname><given-names>J.</given-names></name><name><surname>Zaleski</surname><given-names>C.</given-names></name><name><surname>Jha</surname><given-names>S.</given-names></name><name><surname>Batut</surname><given-names>P.</given-names></name><name><surname>Chaisson</surname><given-names>M.</given-names></name><name><surname>Gingeras</surname><given-names>T.R.</given-names></name></person-group><article-title>STAR: ultrafast universal RNA-seq aligner</article-title><source>Bioinformatics</source><volume>29</volume><year>2013</year><fpage>15</fpage><lpage>21</lpage><pub-id pub-id-type="pmid">23104886</pub-id></element-citation></ref><ref id="bib10"><element-citation publication-type="journal" id="sref10"><person-group person-group-type="author"><name><surname>Downes</surname><given-names>C.S.</given-names></name><name><surname>Clarke</surname><given-names>D.J.</given-names></name><name><surname>Mullinger</surname><given-names>A.M.</given-names></name><name><surname>Giménez-Abián</surname><given-names>J.F.</given-names></name><name><surname>Creighton</surname><given-names>A.M.</given-names></name><name><surname>Johnson</surname><given-names>R.T.</given-names></name></person-group><article-title>A topoisomerase II-dependent G2 cycle checkpoint in mammalian cells/</article-title><source>Nature</source><volume>372</volume><year>1994</year><fpage>467</fpage><lpage>470</lpage><pub-id pub-id-type="pmid">7984241</pub-id></element-citation></ref><ref id="bib11"><element-citation publication-type="journal" id="sref11"><person-group person-group-type="author"><name><surname>Geurts</surname><given-names>P.</given-names></name><name><surname>Ernst</surname><given-names>D.</given-names></name><name><surname>Wehenkel</surname><given-names>L.</given-names></name></person-group><article-title>Extremely randomized trees</article-title><source>Mach. Learn.</source><volume>63</volume><year>2006</year><fpage>3</fpage><lpage>42</lpage></element-citation></ref><ref id="bib12"><element-citation publication-type="journal" id="sref12"><person-group person-group-type="author"><name><surname>Greig</surname><given-names>K.T.</given-names></name><name><surname>Carotta</surname><given-names>S.</given-names></name><name><surname>Nutt</surname><given-names>S.L.</given-names></name></person-group><article-title>Critical roles for c-Myb in hematopoietic progenitor cells</article-title><source>Semin. Immunol.</source><volume>20</volume><year>2008</year><fpage>247</fpage><lpage>256</lpage><pub-id pub-id-type="pmid">18585056</pub-id></element-citation></ref><ref id="bib13"><element-citation publication-type="journal" id="sref13"><person-group person-group-type="author"><name><surname>Guo</surname><given-names>G.</given-names></name><name><surname>Luc</surname><given-names>S.</given-names></name><name><surname>Marco</surname><given-names>E.</given-names></name><name><surname>Lin</surname><given-names>T.-W.</given-names></name><name><surname>Peng</surname><given-names>C.</given-names></name><name><surname>Kerenyi</surname><given-names>M.A.</given-names></name><name><surname>Beyaz</surname><given-names>S.</given-names></name><name><surname>Kim</surname><given-names>W.</given-names></name><name><surname>Xu</surname><given-names>J.</given-names></name><name><surname>Das</surname><given-names>P.P.</given-names></name></person-group><article-title>Mapping cellular hierarchy by single-cell analysis of the cell surface repertoire</article-title><source>Cell Stem Cell</source><volume>13</volume><year>2013</year><fpage>492</fpage><lpage>505</lpage><pub-id pub-id-type="pmid">24035353</pub-id></element-citation></ref><ref id="bib14"><element-citation publication-type="journal" id="sref14"><person-group person-group-type="author"><name><surname>Hensman</surname><given-names>J.</given-names></name><name><surname>Rattray</surname><given-names>M.</given-names></name><name><surname>Lawrence</surname><given-names>N.D.</given-names></name></person-group><article-title>Fast Nonparametric Clustering of Structured Time-Series</article-title><source>IEEE Trans. Pattern Anal. Mach. Intell.</source><volume>37</volume><year>2015</year><fpage>383</fpage><lpage>393</lpage><pub-id pub-id-type="pmid">26353249</pub-id></element-citation></ref><ref id="bib15"><element-citation publication-type="journal" id="sref15"><person-group person-group-type="author"><name><surname>Howe</surname><given-names>K.</given-names></name><name><surname>Clark</surname><given-names>M.D.</given-names></name><name><surname>Torroja</surname><given-names>C.F.</given-names></name><name><surname>Torrance</surname><given-names>J.</given-names></name><name><surname>Berthelot</surname><given-names>C.</given-names></name><name><surname>Muffato</surname><given-names>M.</given-names></name><name><surname>Collins</surname><given-names>J.E.</given-names></name><name><surname>Humphray</surname><given-names>S.</given-names></name><name><surname>McLaren</surname><given-names>K.</given-names></name><name><surname>Matthews</surname><given-names>L.</given-names></name></person-group><article-title>The zebrafish reference genome sequence and its relationship to the human genome</article-title><source>Nature</source><volume>496</volume><year>2013</year><fpage>498</fpage><lpage>503</lpage><pub-id pub-id-type="pmid">23594743</pub-id></element-citation></ref><ref id="bib16"><element-citation publication-type="journal" id="sref16"><person-group person-group-type="author"><name><surname>Hsia</surname><given-names>N.</given-names></name><name><surname>Zon</surname><given-names>L.I.</given-names></name></person-group><article-title>Transcriptional regulation of hematopoietic stem cell development in zebrafish</article-title><source>Exp. Hematol.</source><volume>33</volume><year>2005</year><fpage>1007</fpage><lpage>1014</lpage><pub-id pub-id-type="pmid">16140148</pub-id></element-citation></ref><ref id="bib17"><element-citation publication-type="journal" id="sref17"><person-group person-group-type="author"><name><surname>Hyvärinen</surname><given-names>A.</given-names></name><name><surname>Oja</surname><given-names>E.</given-names></name></person-group><article-title>Independent component analysis: algorithms and applications</article-title><source>Neural Netw.</source><volume>13</volume><year>2000</year><fpage>411</fpage><lpage>430</lpage><pub-id pub-id-type="pmid">10946390</pub-id></element-citation></ref><ref id="bib18"><element-citation publication-type="journal" id="sref18"><person-group person-group-type="author"><name><surname>Jagannathan-Bogdan</surname><given-names>M.</given-names></name><name><surname>Zon</surname><given-names>L.I.</given-names></name></person-group><article-title>Hematopoiesis</article-title><source>Development</source><volume>140</volume><year>2013</year><fpage>2463</fpage><lpage>2467</lpage><pub-id pub-id-type="pmid">23715539</pub-id></element-citation></ref><ref id="bib19"><element-citation publication-type="journal" id="sref19"><person-group person-group-type="author"><name><surname>Li</surname><given-names>B.</given-names></name><name><surname>Dewey</surname><given-names>C.N.</given-names></name></person-group><article-title>RSEM: accurate transcript quantification from RNA-seq data with or without a reference genome</article-title><source>BMC Bioinformatics</source><volume>12</volume><year>2011</year><fpage>323</fpage><pub-id pub-id-type="pmid">21816040</pub-id></element-citation></ref><ref id="bib20"><element-citation publication-type="journal" id="sref20"><person-group person-group-type="author"><name><surname>Loughran</surname><given-names>S.J.</given-names></name><name><surname>Kruse</surname><given-names>E.A.</given-names></name><name><surname>Hacking</surname><given-names>D.F.</given-names></name><name><surname>de Graaf</surname><given-names>C.A.</given-names></name><name><surname>Hyland</surname><given-names>C.D.</given-names></name><name><surname>Willson</surname><given-names>T.A.</given-names></name><name><surname>Henley</surname><given-names>K.J.</given-names></name><name><surname>Ellis</surname><given-names>S.</given-names></name><name><surname>Voss</surname><given-names>A.K.</given-names></name><name><surname>Metcalf</surname><given-names>D.</given-names></name></person-group><article-title>The transcription factor Erg is essential for definitive hematopoiesis and the function of adult hematopoietic stem cells</article-title><source>Nat. Immunol.</source><volume>9</volume><year>2008</year><fpage>810</fpage><lpage>819</lpage><pub-id pub-id-type="pmid">18500345</pub-id></element-citation></ref><ref id="bib21"><element-citation publication-type="journal" id="sref21"><person-group person-group-type="author"><name><surname>Ma</surname><given-names>D.</given-names></name><name><surname>Zhang</surname><given-names>J.</given-names></name><name><surname>Lin</surname><given-names>H.-F.</given-names></name><name><surname>Italiano</surname><given-names>J.</given-names></name><name><surname>Handin</surname><given-names>R.I.</given-names></name></person-group><article-title>The identification and characterization of zebrafish hematopoietic stem cells</article-title><source>Blood</source><volume>118</volume><year>2011</year><fpage>289</fpage><lpage>297</lpage><pub-id pub-id-type="pmid">21586750</pub-id></element-citation></ref><ref id="bib22"><element-citation publication-type="journal" id="sref22"><person-group person-group-type="author"><name><surname>Meyer</surname><given-names>A.</given-names></name><name><surname>Schartl</surname><given-names>M.</given-names></name></person-group><article-title>Gene and genome duplications in vertebrates: the one-to-four (-to-eight in fish) rule and the evolution of novel gene functions</article-title><source>Curr. Opin. Cell Biol.</source><volume>11</volume><year>1999</year><fpage>699</fpage><lpage>704</lpage><pub-id pub-id-type="pmid">10600714</pub-id></element-citation></ref><ref id="bib23"><element-citation publication-type="journal" id="sref23"><person-group person-group-type="author"><name><surname>Muller-Sieburg</surname><given-names>C.E.</given-names></name><name><surname>Sieburg</surname><given-names>H.B.</given-names></name><name><surname>Bernitz</surname><given-names>J.M.</given-names></name><name><surname>Cattarossi</surname><given-names>G.</given-names></name></person-group><article-title>Stem cell heterogeneity: implications for aging and regenerative medicine</article-title><source>Blood</source><volume>119</volume><year>2012</year><fpage>3900</fpage><lpage>3907</lpage><pub-id pub-id-type="pmid">22408258</pub-id></element-citation></ref><ref id="bib24"><element-citation publication-type="journal" id="sref24"><person-group person-group-type="author"><name><surname>Notta</surname><given-names>F.</given-names></name><name><surname>Zandi</surname><given-names>S.</given-names></name><name><surname>Takayama</surname><given-names>N.</given-names></name><name><surname>Dobson</surname><given-names>S.</given-names></name><name><surname>Gan</surname><given-names>O.I.</given-names></name><name><surname>Wilson</surname><given-names>G.</given-names></name><name><surname>Kaufmann</surname><given-names>K.B.</given-names></name><name><surname>McLeod</surname><given-names>J.</given-names></name><name><surname>Laurenti</surname><given-names>E.</given-names></name><name><surname>Dunant</surname><given-names>C.F.</given-names></name></person-group><article-title>Distinct routes of lineage development reshape the human blood hierarchy across ontogeny</article-title><source>Science</source><year>2015</year><fpage>aab2116</fpage><pub-id pub-id-type="pmid">26541609</pub-id></element-citation></ref><ref id="bib25"><element-citation publication-type="journal" id="sref25"><person-group person-group-type="author"><name><surname>Orkin</surname><given-names>S.H.</given-names></name><name><surname>Zon</surname><given-names>L.I.</given-names></name></person-group><article-title>Hematopoiesis: an evolving paradigm for stem cell biology</article-title><source>Cell</source><volume>132</volume><year>2008</year><fpage>631</fpage><lpage>644</lpage><pub-id pub-id-type="pmid">18295580</pub-id></element-citation></ref><ref id="bib26"><element-citation publication-type="journal" id="sref26"><person-group person-group-type="author"><name><surname>Patro</surname><given-names>R.</given-names></name><name><surname>Mount</surname><given-names>S.M.</given-names></name><name><surname>Kingsford</surname><given-names>C.</given-names></name></person-group><article-title>Sailfish enables alignment-free isoform quantification from RNA-seq reads using lightweight algorithms</article-title><source>Nat. Biotechnol.</source><volume>32</volume><year>2014</year><fpage>462</fpage><lpage>464</lpage><pub-id pub-id-type="pmid">24752080</pub-id></element-citation></ref><ref id="bib27"><mixed-citation publication-type="other" id="sref27">Patro, R., Duggal, G., Kingsford, C., 2015. Salmon: accurate, versatile and ultrafast quantification from RNA-seq data using lightweight-alignment. Published online June 27, 2015. <ext-link ext-link-type="uri" xlink:href="http://dx.doi.org/10.1101/021592" id="intref0020">http://dx.doi.org/10.1101/021592</ext-link>.</mixed-citation></ref><ref id="bib28"><element-citation publication-type="journal" id="sref28"><person-group person-group-type="author"><name><surname>Pedregosa</surname><given-names>F.</given-names></name><name><surname>Varoquaux</surname><given-names>G.</given-names></name><name><surname>Gramfort</surname><given-names>A.</given-names></name><name><surname>Michel</surname><given-names>V.</given-names></name><name><surname>Thirion</surname><given-names>B.</given-names></name><name><surname>Grisel</surname><given-names>O.</given-names></name><name><surname>Blondel</surname><given-names>M.</given-names></name><name><surname>Prettenhofer</surname><given-names>P.</given-names></name><name><surname>Weiss</surname><given-names>R.</given-names></name><name><surname>Dubourg</surname><given-names>V.</given-names></name></person-group><article-title>Scikit-learn: machine learning in python</article-title><source>J. Mach. Learn. Res.</source><volume>12</volume><year>2011</year><fpage>2825</fpage><lpage>2830</lpage></element-citation></ref><ref id="bib29"><element-citation publication-type="journal" id="sref29"><person-group person-group-type="author"><name><surname>Picelli</surname><given-names>S.</given-names></name><name><surname>Björklund</surname><given-names>Å.K.</given-names></name><name><surname>Faridani</surname><given-names>O.R.</given-names></name><name><surname>Sagasser</surname><given-names>S.</given-names></name><name><surname>Winberg</surname><given-names>G.</given-names></name><name><surname>Sandberg</surname><given-names>R.</given-names></name></person-group><article-title>Smart-seq2 for sensitive full-length transcriptome profiling in single cells</article-title><source>Nat. Methods</source><volume>10</volume><year>2013</year><fpage>1096</fpage><lpage>1098</lpage><pub-id pub-id-type="pmid">24056875</pub-id></element-citation></ref><ref id="bib30"><element-citation publication-type="journal" id="sref30"><person-group person-group-type="author"><name><surname>Picelli</surname><given-names>S.</given-names></name><name><surname>Faridani</surname><given-names>O.R.</given-names></name><name><surname>Björklund</surname><given-names>A.K.</given-names></name><name><surname>Winberg</surname><given-names>G.</given-names></name><name><surname>Sagasser</surname><given-names>S.</given-names></name><name><surname>Sandberg</surname><given-names>R.</given-names></name></person-group><article-title>Full-length RNA-seq from single cells using Smart-seq2</article-title><source>Nat. Protoc.</source><volume>9</volume><year>2014</year><fpage>171</fpage><lpage>181</lpage><pub-id pub-id-type="pmid">24385147</pub-id></element-citation></ref><ref id="bib31"><element-citation publication-type="journal" id="sref31"><person-group person-group-type="author"><name><surname>Pineault</surname><given-names>N.</given-names></name><name><surname>Helgason</surname><given-names>C.D.</given-names></name><name><surname>Lawrence</surname><given-names>H.J.</given-names></name><name><surname>Humphries</surname><given-names>R.K.</given-names></name></person-group><article-title>Differential expression of Hox, Meis1, and Pbx1 genes in primitive cells throughout murine hematopoietic ontogeny</article-title><source>Exp. Hematol.</source><volume>30</volume><year>2002</year><fpage>49</fpage><lpage>57</lpage><pub-id pub-id-type="pmid">11823037</pub-id></element-citation></ref><ref id="bib32"><element-citation publication-type="journal" id="sref32"><person-group person-group-type="author"><name><surname>Poirault-Chassac</surname><given-names>S.</given-names></name><name><surname>Six</surname><given-names>E.</given-names></name><name><surname>Catelain</surname><given-names>C.</given-names></name><name><surname>Lavergne</surname><given-names>M.</given-names></name><name><surname>Villeval</surname><given-names>J.-L.</given-names></name><name><surname>Vainchenker</surname><given-names>W.</given-names></name><name><surname>Lauret</surname><given-names>E.</given-names></name></person-group><article-title>Notch/Delta4 signaling inhibits human megakaryocytic terminal differentiation</article-title><source>Blood</source><volume>116</volume><year>2010</year><fpage>5670</fpage><lpage>5678</lpage><pub-id pub-id-type="pmid">20829371</pub-id></element-citation></ref><ref id="bib33"><element-citation publication-type="journal" id="sref33"><person-group person-group-type="author"><name><surname>Reimand</surname><given-names>J.</given-names></name><name><surname>Arak</surname><given-names>T.</given-names></name><name><surname>Vilo</surname><given-names>J.</given-names></name></person-group><article-title>g:Profiler--a web server for functional interpretation of gene lists (2011 update)</article-title><source>Nucleic Acids Res.</source><volume>39</volume><year>2011</year><fpage>W307</fpage><lpage>W315</lpage><pub-id pub-id-type="pmid">21646343</pub-id></element-citation></ref><ref id="bib34"><element-citation publication-type="journal" id="sref34"><person-group person-group-type="author"><name><surname>Robin</surname><given-names>C.</given-names></name><name><surname>Ottersbach</surname><given-names>K.</given-names></name><name><surname>Boisset</surname><given-names>J.-C.</given-names></name><name><surname>Oziemlak</surname><given-names>A.</given-names></name><name><surname>Dzierzak</surname><given-names>E.</given-names></name></person-group><article-title>CD41 is developmentally regulated and differentially expressed on mouse hematopoietic stem cells</article-title><source>Blood</source><volume>117</volume><year>2011</year><fpage>5088</fpage><lpage>5091</lpage><pub-id pub-id-type="pmid">21415271</pub-id></element-citation></ref><ref id="bib35"><element-citation publication-type="journal" id="sref35"><person-group person-group-type="author"><name><surname>Sanjuan-Pla</surname><given-names>A.</given-names></name><name><surname>Macaulay</surname><given-names>I.C.</given-names></name><name><surname>Jensen</surname><given-names>C.T.</given-names></name><name><surname>Woll</surname><given-names>P.S.</given-names></name><name><surname>Luis</surname><given-names>T.C.</given-names></name><name><surname>Mead</surname><given-names>A.</given-names></name><name><surname>Moore</surname><given-names>S.</given-names></name><name><surname>Carella</surname><given-names>C.</given-names></name><name><surname>Matsuoka</surname><given-names>S.</given-names></name><name><surname>Bouriez Jones</surname><given-names>T.</given-names></name></person-group><article-title>Platelet-biased stem cells reside at the apex of the haematopoietic stem-cell hierarchy</article-title><source>Nature</source><volume>502</volume><year>2013</year><fpage>232</fpage><lpage>236</lpage><pub-id pub-id-type="pmid">23934107</pub-id></element-citation></ref><ref id="bib36"><element-citation publication-type="journal" id="sref36"><person-group person-group-type="author"><name><surname>Schick</surname><given-names>P.K.</given-names></name><name><surname>Konkle</surname><given-names>B.A.</given-names></name><name><surname>He</surname><given-names>X.</given-names></name><name><surname>Thornton</surname><given-names>R.D.</given-names></name></person-group><article-title>P-selectin mRNA is expressed at a later phase of megakaryocyte maturation than mRNAs for von Willebrand factor and glycoprotein Ib-alpha</article-title><source>J. Lab. Clin. Med.</source><volume>121</volume><year>1993</year><fpage>714</fpage><lpage>721</lpage><pub-id pub-id-type="pmid">7683032</pub-id></element-citation></ref><ref id="bib37"><element-citation publication-type="journal" id="sref37"><person-group person-group-type="author"><name><surname>Schulte</surname><given-names>R.</given-names></name><name><surname>Wilson</surname><given-names>N.K.</given-names></name><name><surname>Prick</surname><given-names>J.C.M.</given-names></name><name><surname>Cossetti</surname><given-names>C.</given-names></name><name><surname>Maj</surname><given-names>M.K.</given-names></name><name><surname>Gottgens</surname><given-names>B.</given-names></name><name><surname>Kent</surname><given-names>D.G.</given-names></name></person-group><article-title>Index sorting resolves heterogeneous murine hematopoietic stem cell populations</article-title><source>Exp. Hematol.</source><volume>43</volume><year>2015</year><fpage>803</fpage><lpage>811</lpage><pub-id pub-id-type="pmid">26051918</pub-id></element-citation></ref><ref id="bib38"><element-citation publication-type="journal" id="sref38"><person-group person-group-type="author"><name><surname>Seita</surname><given-names>J.</given-names></name><name><surname>Weissman</surname><given-names>I.L.</given-names></name></person-group><article-title>Hematopoietic stem cell: self-renewal versus differentiation</article-title><source>Wiley Interdiscip. Rev. Syst. Biol. Med.</source><volume>2</volume><year>2010</year><fpage>640</fpage><lpage>653</lpage><pub-id pub-id-type="pmid">20890962</pub-id></element-citation></ref><ref id="bib39"><element-citation publication-type="journal" id="sref39"><person-group person-group-type="author"><name><surname>Song</surname><given-names>H.-D.</given-names></name><name><surname>Sun</surname><given-names>X.-J.</given-names></name><name><surname>Deng</surname><given-names>M.</given-names></name><name><surname>Zhang</surname><given-names>G.-W.</given-names></name><name><surname>Zhou</surname><given-names>Y.</given-names></name><name><surname>Wu</surname><given-names>X.-Y.</given-names></name><name><surname>Sheng</surname><given-names>Y.</given-names></name><name><surname>Chen</surname><given-names>Y.</given-names></name><name><surname>Ruan</surname><given-names>Z.</given-names></name><name><surname>Jiang</surname><given-names>C.-L.</given-names></name></person-group><article-title>Hematopoietic gene expression profile in zebrafish kidney marrow</article-title><source>Proc. Natl. Acad. Sci. USA</source><volume>101</volume><year>2004</year><fpage>16240</fpage><lpage>16245</lpage><pub-id pub-id-type="pmid">15520368</pub-id></element-citation></ref><ref id="bib40"><element-citation publication-type="journal" id="sref40"><person-group person-group-type="author"><name><surname>Stachura</surname><given-names>D.L.</given-names></name><name><surname>Reyes</surname><given-names>J.R.</given-names></name><name><surname>Bartunek</surname><given-names>P.</given-names></name><name><surname>Paw</surname><given-names>B.H.</given-names></name><name><surname>Zon</surname><given-names>L.I.</given-names></name><name><surname>Traver</surname><given-names>D.</given-names></name></person-group><article-title>Zebrafish kidney stromal cell lines support multilineage hematopoiesis</article-title><source>Blood</source><volume>114</volume><year>2009</year><fpage>279</fpage><lpage>289</lpage><pub-id pub-id-type="pmid">19433857</pub-id></element-citation></ref><ref id="bib41"><element-citation publication-type="journal" id="sref41"><person-group person-group-type="author"><name><surname>Tenen</surname><given-names>D.G.</given-names></name><name><surname>Hromas</surname><given-names>R.</given-names></name><name><surname>Licht</surname><given-names>J.D.</given-names></name><name><surname>Zhang</surname><given-names>D.E.</given-names></name></person-group><article-title>Transcription factors, normal myeloid development, and leukemia</article-title><source>Blood</source><volume>90</volume><year>1997</year><fpage>489</fpage><lpage>519</lpage><pub-id pub-id-type="pmid">9226149</pub-id></element-citation></ref><ref id="bib42"><mixed-citation publication-type="other" id="sref42">The GPy authors, n.d. GPy: A Gaussian process framework in python. <ext-link ext-link-type="uri" xlink:href="http://github.com/SheffieldML/GPy" id="intref0025">http://github.com/SheffieldML/GPy</ext-link>.</mixed-citation></ref><ref id="bib43"><mixed-citation publication-type="other" id="sref43">Titsias, M.K., and Lawrence, N.D. 2010. Bayesian Gaussian process latent variable model. In Proceedings of the 13th International Conference on Artificial Intelligence and Statistics, pp. 844–851.</mixed-citation></ref><ref id="bib44"><element-citation publication-type="journal" id="sref44"><person-group person-group-type="author"><name><surname>Van der Maaten</surname><given-names>L.</given-names></name><name><surname>Hinton</surname><given-names>G.</given-names></name></person-group><article-title>Visualizing data using t-SNE</article-title><source>J. Mach. Learn. Res.</source><volume>9</volume><year>2008</year><fpage>85</fpage></element-citation></ref><ref id="bib45"><element-citation publication-type="journal" id="sref45"><person-group person-group-type="author"><name><surname>Vassen</surname><given-names>L.</given-names></name><name><surname>Okayama</surname><given-names>T.</given-names></name><name><surname>Möröy</surname><given-names>T.</given-names></name></person-group><article-title>Gfi1b:green fluorescent protein knock-in mice reveal a dynamic expression pattern of Gfi1b during hematopoiesis that is largely complementary to Gfi1</article-title><source>Blood</source><volume>109</volume><year>2007</year><fpage>2356</fpage><lpage>2364</lpage><pub-id pub-id-type="pmid">17095621</pub-id></element-citation></ref><ref id="bib46"><element-citation publication-type="journal" id="sref46"><person-group person-group-type="author"><name><surname>Wagner</surname><given-names>G.P.</given-names></name><name><surname>Kin</surname><given-names>K.</given-names></name><name><surname>Lynch</surname><given-names>V.J.</given-names></name></person-group><article-title>Measurement of mRNA abundance using RNA-seq data: RPKM measure is inconsistent among samples</article-title><source>Theory Biosci.</source><volume>131</volume><year>2012</year><fpage>281</fpage><lpage>285</lpage><pub-id pub-id-type="pmid">22872506</pub-id></element-citation></ref><ref id="bib47"><element-citation publication-type="journal" id="sref47"><person-group person-group-type="author"><name><surname>Wang</surname><given-names>L.</given-names></name><name><surname>Wang</surname><given-names>S.</given-names></name><name><surname>Li</surname><given-names>W.</given-names></name></person-group><article-title>RSeQC: quality control of RNA-seq experiments</article-title><source>Bioinformatics</source><volume>28</volume><year>2012</year><fpage>2184</fpage><lpage>2185</lpage><pub-id pub-id-type="pmid">22743226</pub-id></element-citation></ref><ref id="bib48"><element-citation publication-type="journal" id="sref48"><person-group person-group-type="author"><name><surname>Ward</surname><given-names>J.H.</given-names><suffix>Jr.</suffix></name></person-group><article-title>Hierarchical grouping to optimize an objective function</article-title><source>J. Am. Stat. Assoc.</source><volume>58</volume><year>1963</year><fpage>236</fpage><lpage>244</lpage></element-citation></ref><ref id="bib49"><element-citation publication-type="journal" id="sref49"><person-group person-group-type="author"><name><surname>Wright</surname><given-names>D.E.</given-names></name><name><surname>Bowman</surname><given-names>E.P.</given-names></name><name><surname>Wagers</surname><given-names>A.J.</given-names></name><name><surname>Butcher</surname><given-names>E.C.</given-names></name><name><surname>Weissman</surname><given-names>I.L.</given-names></name></person-group><article-title>Hematopoietic stem cells are uniquely selective in their migratory response to chemokines</article-title><source>J. Exp. Med.</source><volume>195</volume><year>2002</year><fpage>1145</fpage><lpage>1154</lpage><pub-id pub-id-type="pmid">11994419</pub-id></element-citation></ref><ref id="bib50"><element-citation publication-type="journal" id="sref50"><person-group person-group-type="author"><name><surname>Zeng</surname><given-names>H.</given-names></name><name><surname>Yücel</surname><given-names>R.</given-names></name><name><surname>Kosan</surname><given-names>C.</given-names></name><name><surname>Klein-Hitpass</surname><given-names>L.</given-names></name><name><surname>Möröy</surname><given-names>T.</given-names></name></person-group><article-title>Transcription factor Gfi1 regulates self-renewal and engraftment of hematopoietic stem cells</article-title><source>EMBO J.</source><volume>23</volume><year>2004</year><fpage>4116</fpage><lpage>4125</lpage><pub-id pub-id-type="pmid">15385956</pub-id></element-citation></ref></ref-list><sec id="app1"><title>Accession Numbers</title><p>The accession number for the data reported in this paper is ArrayExpress: <ext-link ext-link-type="uri" xlink:href="array-express:E-MTAB-3947" id="intref0010">E-MTAB-3947</ext-link>.</p></sec><sec id="app3" sec-type="supplementary-material"><title>Supplemental Information</title><p><supplementary-material content-type="local-data" id="mmc1"><caption><title>Document S1. Supplemental Figures S1–S7</title></caption><media xlink:href="mmc1.pdf"></media></supplementary-material><supplementary-material content-type="local-data" id="mmc2"><caption><title>Table S1. Top Genes Distinguishing Each Cluster Based on Random Forest Feature Importance, Related to Figure 4</title></caption><media xlink:href="mmc2.xlsx"></media></supplementary-material><supplementary-material content-type="local-data" id="mmc3"><caption><title>Table S2. List of Genes with Dynamic Expression over Pseudotime, Annotated by which Temporal Expression Group They Belong To, Related to Figure 6</title><p>Together with enriched functional gene sets for the different temporal expression groups.</p></caption><media xlink:href="mmc3.xlsx"></media></supplementary-material><supplementary-material content-type="local-data" id="mmc4"><caption><title>Table S3. List of Expressed Ohnolog Gene Pairs Annotated by the Decision Tree Classification into Single, Mixed, and XOR -Ohnologs, Related to Figure 7</title></caption><media xlink:href="mmc4.xlsx"></media></supplementary-material><supplementary-material content-type="local-data" id="mmc5"><caption><title>Data S1. Sample Information, Related to Figures 1, 2, 3, 4, 5, 6, and 7</title><p>This table contains detailed information about each sample, which was inferred by analysis and used to create most figure panels.</p></caption><media xlink:href="mmc5.zip"></media></supplementary-material><supplementary-material content-type="local-data" id="mmc6"><caption><title>Data S2. Analysis Files, Related to Figures 1, 2, 3, 4, 5, 6, and 7</title><p>This contains the scripts, in the form of IPython notebooks, to reproduce all analysis and most of the figure panels in the text.</p></caption><media xlink:href="mmc6.zip"></media></supplementary-material><supplementary-material content-type="local-data" id="mmc7"><caption><title>Document S2. Article plus Supplemental Information</title></caption><media xlink:href="mmc7.pdf"></media></supplementary-material></p></sec><ack id="ack0010"><title>Acknowledgments</title><p>We thank the Sanger-EBI single cell centre and Chiara Cossetti, Michal Maj, and Reiner Schulte at the CIMR flow cytometry Core for their help with index cell sorting. The study was supported by Cancer Research UK grant number C45041/A14953 (to A.C., C.L., and L.F.) and a core support grant from the Wellcome Trust and MRC to the Wellcome Trust – Medical Research Council Cambridge Stem Cell Institute. T.V. acknowledges the Wellcome Trust and KU Leuven (SymBioSys, PFV/10/016). S.T would like to acknowledge the Lister Research Prize from the Lister Institute.</p></ack><fn-group><fn id="d32e206"><p id="ccnp0005">This is an open access article under the CC BY license (<ext-link ext-link-type="uri" xlink:href="http://creativecommons.org/licenses/by/4.0/" id="ccintref0005">http://creativecommons.org/licenses/by/4.0/</ext-link>).</p></fn><fn id="app2" fn-type="supplementary-material"><p>Supplemental Information includes seven figures, three tables, and two data files and can be found with this article online at <ext-link ext-link-type="doi" xlink:href="10.1016/j.celrep.2015.12.082" id="intref0015">http://dx.doi.org/10.1016/j.celrep.2015.12.082</ext-link>.</p></fn></fn-group></back><floats-group><fig id="fig1"><label>Figure 1</label><caption><p><italic>cd41</italic> Cells Transition through Five Transcriptional States during Thrombocyte Differentiation in Zebrafish</p><p>(A) A single kidney, from a heterozygote <italic>Tg(cd41:EGFP)</italic> reporter fish, was dissected and carefully passed through a strainer. Using flow cytometry, EGFP<sup>low</sup> and EGFP<sup>high</sup> cells were identified and 188 cells from each population were index sorted. Two wells (in red) per plate were left without cells. RNA from each cell was isolated and used to construct a single mRNA-seq library per cell, which was then sequenced using Hi-seq.</p><p>(B) t-SNE plot of the RNA-seq data from 363 EGFP<sup>low</sup> and EGFP<sup>high</sup> cells.</p><p>(C) The same t-SNE plot (as shown in B) but with points colored based on the cluster the cells belong to. Clusters are labeled as 1a, 1b, 2, 3, 4, and outlier cells.</p><p>See also <xref rid="mmc1" ref-type="supplementary-material">Figures S1</xref>, <xref rid="mmc1" ref-type="supplementary-material">S2</xref>, and <xref rid="mmc1" ref-type="supplementary-material">S3</xref>.</p></caption><graphic xlink:href="gr1"></graphic></fig><fig id="fig2"><label>Figure 2</label><caption><p>Ordering of Cells through the Developmental Trajectory</p><p>(A) We inferred a smooth progression over the developmental lineage, represented as pseudotime, using two different methods. Here we demonstrate the path of both pseudotimes by regressing them into a t-SNE plot of the data. Points are colored based on the cluster the cells belong to.</p><p>(B) The pseudotime inferred with two different methods correlate very strongly (Spearman correlation 0.97).</p><p>(C) Expression of <italic>cd41</italic> mRNA (top), <italic>GFP</italic> mRNA (middle), and GFP fluorescence (bottom) shown in pseudotime. Each point represents an individual cell; points are colored based on the cluster the cells belong to.</p><p>See also <xref rid="mmc1" ref-type="supplementary-material">Figure S3</xref>.</p></caption><graphic xlink:href="gr2"></graphic></fig><fig id="fig3"><label>Figure 3</label><caption><p>Expression of Key Regulators of Hematopoiesis over Pseudotime</p><p>Expression (in TPM) of genes, relevant in hematopoiesis, over pseudotime. Points are colored based on the cluster the cells belong to. For each cluster, we show the proportion of cells within the given cluster expressing the gene at TPM > 1. HSC, hematopoietic stem cells-affiliated genes; Meg-Erythroid, megakaryocyte-erythroid progenitors-affiliated genes; Myeloid, myeloid lineage-affiliated genes.</p></caption><graphic xlink:href="gr3"></graphic></fig><fig id="fig4"><label>Figure 4</label><caption><p>Identification of New Cell-type Markers</p><p>(A) t-SNE plot of the RNA-seq data from 363 EGFP cells. Points are colored based on the cluster the cells belong to. Selected genes, whose expression is highly correlated with individual clusters, are shown next to each cluster. Selected gene ontology terms associated with genes that are highly correlated with cluster 2 and the outlier cells are included.</p><p>(B) Expression of marker genes over pseudotime (left). Points are colored based on the cluster the cells belong to. For each cluster, we show the proportion of cells expressing the gene at TPM > 1. Expression of pairs of genes is shown on the right. Points are colored based on the cluster the cells belong to. The side diagrams show the proportion of cells within the cluster expressing the gene at the given level of expression.</p><p>(C) Cell cycle analysis of three different populations of EGFP cells. The GFP<sup>low</sup>SSC<sup>high</sup> cells are enriched for cells from clusters 1a/1b/2, GFP<sup>low</sup>SSC<sup>low</sup> and GFP<sup>high</sup> cells are enriched for cells from clusters 3 and 4, respectively. An average of two experiments is shown as a percentage of cells in G0 and G1 (G0/1) and S and G2 phase (S/G2) ± SEM.</p><p>(D) Distribution of FSC (top) and SSC (bottom) values in the different clusters. In particular, one can see that the small population of outliers (cluster x, shaded gray) has higher FSC and SSC values than cells from other clusters.</p><p>See also <xref rid="mmc1" ref-type="supplementary-material">Figures S4</xref> and <xref rid="mmc1" ref-type="supplementary-material">S5</xref>.</p></caption><graphic xlink:href="gr4"></graphic></fig><fig id="fig5"><label>Figure 5</label><caption><p>Validation of Identified Early Clusters, and Terminal State of Late Cluster</p><p>(A–C) In a second experiment, cells only belonging to the early clusters were sorted. The distributions of (A) <italic>cd41</italic> expression, (B) number of expressed genes, and (C) endogenous mRNA content of cells, are as expected in the populations of cells sorted from kidney.</p><p>(D) mRNA expression of <italic>cd41</italic> in the sorted populations of cells. We see the expected increase from Kidney EarlyEnriched through Kidney EGFP<sup>low</sup> to finally Kidney EGFP<sup>high</sup>. Expression of <italic>cd41</italic> did not change between Kidney EGFP<sup>high</sup> and EGFP<sup>high</sup> cells in circulation (likelihood ratio test, p = 1 after correcting for multiple testing.)</p><p>(E and F) When developing from EarlyEnriched through EGFP<sup>low</sup> to EGFP<sup>high</sup>, the cells express fewer genes and contain less mRNA, confirming the pseudotime ordering we observed in the initial experiment. There was no change in the number of expressed genes and RNA content between kidney- and circulation-derived EGFP<sup>high</sup> cells.</p><p>See also <xref rid="mmc1" ref-type="supplementary-material">Figures S6</xref> and <xref rid="mmc1" ref-type="supplementary-material">S7</xref>.</p></caption><graphic xlink:href="gr5"></graphic></fig><fig id="fig6"><label>Figure 6</label><caption><p>Identification of Genes that Are Dynamically Regulated over Pseudotime</p><p>(A) Pseudotime expression patterns of genes (rows) that significantly vary over pseudotime progression (x axis). Every row is the <italic>Z</italic> score scaled Gaussian process representing the expression pattern.</p><p>(B) The gene expression pattern for the underlying function explaining the expression pattern in each group is shown as a black line (95% confidence interval in the gray area). Below, selected gene ontology terms associated with the genes in each group are shown.</p><p>(C) Expression (in TPM) of an example gene from each group through pseudotime. Points are colored based on the cluster the cells belong to.</p></caption><graphic xlink:href="gr6"></graphic></fig><fig id="fig7"><label>Figure 7</label><caption><p>Single Cell Analysis Reveals Three Main Patterns of Usage of Duplicated Genes during Thrombopoiesis in Zebrafish</p><p>(A) Ohnolog gene pairs were divided into four classes based on thresholds in a decision tree.</p><p>(B) Expression (in TPM) of example ohnologs, randomly selected from each class, in individual cells. Points are colored based on the cluster the cells belong to. XOR ohnolog: both ohnologs are expressed but never in the same cell. Single ohnolog: just one ohnolog is expressed. Mixed ohnologs: both ohnologs are expressed in individual cells.</p></caption><graphic xlink:href="gr7"></graphic></fig></floats-group></pmc></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Belgique/explor/OpenAccessBelV2/Data/Pmc/Corpus
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000346 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Pmc/Corpus/biblio.hfd -nk 000346 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Wicri/Belgique
|area= OpenAccessBelV2
|flux= Pmc
|étape= Corpus
|type= RBID
|clé= PMC:4742565
|texte= Single-Cell RNA-Sequencing Reveals a Continuous Spectrum of Differentiation in Hematopoietic Cells
}}
Pour générer des pages wiki
HfdIndexSelect -h $EXPLOR_AREA/Data/Pmc/Corpus/RBID.i -Sk "pubmed:26804912" \
| HfdSelect -Kh $EXPLOR_AREA/Data/Pmc/Corpus/biblio.hfd \
| NlmPubMed2Wicri -a OpenAccessBelV2
| This area was generated with Dilib version V0.6.25. |  |