Single-Cell Transcriptomics Meets Lineage Tracing
2018; Elsevier BV; Volume: 23; Issue: 2 Linguagem: Inglês
10.1016/j.stem.2018.04.014
ISSN1934-5909
AutoresLennart Kester, Alexander van Oudenaarden,
Tópico(s)T-cell and B-cell Immunology
ResumoReconstructing lineage relationships between cells within a tissue or organism is a long-standing aim in biology. Traditionally, lineage tracing has been achieved through the (genetic) labeling of a cell followed by the tracking of its offspring. Currently, lineage trajectories can also be predicted using single-cell transcriptomics. Although single-cell transcriptomics provides detailed phenotypic information, the predicted lineage trajectories do not necessarily reflect genetic relationships. Recently, techniques have been developed that unite these strategies. In this Review, we discuss transcriptome-based lineage trajectory prediction algorithms, single-cell genetic lineage tracing, and the promising combination of these techniques for stem cell and cancer research. Reconstructing lineage relationships between cells within a tissue or organism is a long-standing aim in biology. Traditionally, lineage tracing has been achieved through the (genetic) labeling of a cell followed by the tracking of its offspring. Currently, lineage trajectories can also be predicted using single-cell transcriptomics. Although single-cell transcriptomics provides detailed phenotypic information, the predicted lineage trajectories do not necessarily reflect genetic relationships. Recently, techniques have been developed that unite these strategies. In this Review, we discuss transcriptome-based lineage trajectory prediction algorithms, single-cell genetic lineage tracing, and the promising combination of these techniques for stem cell and cancer research. Understanding the lineage through which tissues and organisms are formed is one of the fundamental questions in biology. Identifying these relationships will provide invaluable information not only on normal tissue development and homeostasis, but also on developmental disorders and pathologies such as cancer. Historically, lineage tracing is accomplished through the introduction of a heritable mark in a cell, followed by the tracking of its progeny. The different cell types that comprise the progeny are developmentally related since all of these marked cells originate from the same founder cell. Furthermore, the variety of cell types found in the progeny represents the potential of the founder cell. In order to accurately predict the potential of the founder cell, lineage tracing requires accurate cell-type identification. Ideally, one would use as many markers as possible to achieve accurate and precise cell-type classification. However, cell-type identification is usually based on a limited number of markers, thereby potentially masking variability within a subpopulation of cells that express the selected marker genes. This approach might therefore give a biased view on organ complexity. Recent advances in single-cell transcriptomics technologies now allow transcriptome profiling of thousands of single cells, giving unprecedented resolution in cell-type identification and deepening our understanding of tissue complexity (Grün et al., 2015Grün D. Lyubimova A. Kester L. Wiebrands K. Basak O. Sasaki N. Clevers H. van Oudenaarden A. Single-cell messenger RNA sequencing reveals rare intestinal cell types.Nature. 2015; 525: 251-255Crossref PubMed Scopus (286) Google Scholar, Haber et al., 2017Haber A.L. Biton M. Rogel N. Herbst R.H. Shekhar K. Smillie C. Burgin G. Delorey T.M. Howitt M.R. Katz Y. et al.A single-cell survey of the small intestinal epithelium.Nature. 2017; 551: 333-339Crossref PubMed Scopus (27) Google Scholar, Jaitin et al., 2014Jaitin D.A. Kenigsberg E. Keren-Shaul H. Elefant N. Paul F. Zaretsky I. Mildner A. Cohen N. Jung S. Tanay A. Amit I. Massively parallel single-cell RNA-seq for marker-free decomposition of tissues into cell types.Science. 2014; 343: 776-779Crossref PubMed Scopus (460) Google Scholar, Klein et al., 2015Klein A.M. Mazutis L. Akartuna I. Tallapragada N. Veres A. Li V. Peshkin L. Weitz D.A. Kirschner M.W. Droplet barcoding for single-cell transcriptomics applied to embryonic stem cells.Cell. 2015; 161: 1187-1201Abstract Full Text Full Text PDF PubMed Scopus (538) Google Scholar, Macosko et al., 2015Macosko E.Z. Basu A. Satija R. Nemesh J. Shekhar K. Goldman M. Tirosh I. Bialas A.R. Kamitaki N. Martersteck E.M. et al.Highly parallel genome-wide expression profiling of individual cells using nanoliter droplets.Cell. 2015; 161: 1202-1214Abstract Full Text Full Text PDF PubMed Scopus (920) Google Scholar, Zeisel et al., 2015Zeisel A. Muñoz-Manchado A.B. Codeluppi S. Lönnerberg P. La Manno G. Juréus A. Marques S. Munguba H. He L. Betsholtz C. et al.Brain structure. Cell types in the mouse cortex and hippocampus revealed by single-cell RNA-seq.Science. 2015; 347: 1138-1142Crossref PubMed Scopus (641) Google Scholar). The power of single-cell transcriptomics has led researchers to start large-scale sequencing projects such as “The Human Cell Atlas” and “Fly Cell Atlas,” endeavors aimed toward sequencing all cell types present in the human body (Rozenblatt-Rosen et al., 2017Rozenblatt-Rosen O. Stubbington M.J.T. Regev A. Teichmann S.A. The Human Cell Atlas: From vision to reality.Nature. 2017; 550: 451-453Crossref PubMed Scopus (11) Google Scholar) and fruit fly, respectively. In addition, the NIH Brain Initiative (Insel et al., 2013Insel T.R. Landis S.C. Collins F.S. The NIH BRAIN InitiativeResearch priorities.Science. 2013; 340: 687-688Crossref PubMed Scopus (127) Google Scholar) is funding projects aimed at sequencing all cell types present in the human and rodent brain. In parallel, there have been considerable advances in computational methods aimed at performing lineage trajectory reconstruction based on single-cell transcriptomics data (Cannoodt et al., 2016Cannoodt R. Saelens W. Saeys Y. Computational methods for trajectory inference from single-cell transcriptomics.Eur. J. Immunol. 2016; 46: 2496-2506Crossref PubMed Scopus (2) Google Scholar), allowing researchers to sort the transcriptomes of single cells according to their differentiation status. Since lineage reconstruction based on single-cell transcriptomics is independent from the true genetic relationship between cells, we reserve the term “lineage tracing” for genetic lineage tracing and use the term “differentiation trajectories” for transcriptome-derived lineage predictions. However, new experimental techniques that combine single-cell transcriptome sequencing with genetic lineage labels provide information on the relationships between cells for lineage reconstruction along with detailed phenotypic information (Alemany et al., 2018Alemany A. Florescu M. Baron C.S. Peterson-Maduro J. van Oudenaarden A. Whole-organism clone tracing using single-cell sequencing.Nature. 2018; 556: 108-112Crossref PubMed Scopus (7) Google Scholar, Frieda et al., 2017Frieda K.L. Linton J.M. Hormoz S. Choi J. Chow K.K. Singer Z.S. Budde M.W. Elowitz M.B. Cai L. Synthetic recording and in situ readout of lineage information in single cells.Nature. 2017; 541: 107-111Crossref PubMed Scopus (55) Google Scholar, Raj et al., 2018Raj B. Wagner D.E. McKenna A. Pandey S. Klein A.M. Shendure J. Gagnon J.A. Schier A.F. Simultaneous single-cell profiling of lineages and cell types in the vertebrate brain.Nat. Biotechnol. 2018; Crossref PubMed Google Scholar, Spanjaard et al., 2018Spanjaard B. Hu B. Mitic N. Olivares-Chauvet P. Janjuha S. Ninov N. Junker J.P. Simultaneous lineage tracing and cell-type identification using CRISPR-Cas9-induced genetic scars.Nat. Biotechnol. 2018; (Published online April 9, 2018)https://doi.org/10.1038/nbt/4124Crossref PubMed Google Scholar, Yao et al., 2017Yao Z. Mich J.K. Ku S. Menon V. Krostag A.R. Martinez R.A. Furchtgott L. Mulholland H. Bort S. Fuqua M.A. et al.A single-cell roadmap of lineage bifurcation in human esc models of embryonic brain development.Cell Stem Cell. 2017; 20: 120-134Abstract Full Text Full Text PDF PubMed Scopus (23) Google Scholar). This integration of single-cell lineage tracing and transcriptomics will be incredibly powerful, as it allows coarse lineage reconstruction based on genetically heritable marks, followed by refinement based on the transcriptome-derived differentiation trajectories and the assessment of gene-expression changes as a function of the developmental cell state (Figure 1). In this review, we discuss some of the numerous strategies used to predict differentiation trajectories based on single-cell transcriptomics and highlight their applications in various biological systems. Next, we discuss recent developments in prospective lineage tracing that entails the introduction of a heritable mark and retrospective lineage tracing, which exploits naturally occurring elements in the genome (Woodworth et al., 2017Woodworth M.B. Girskis K.M. Walsh C.A. Building a lineage from single cells: Genetic techniques for cell lineage tracking.Nat. Rev. Genet. 2017; 18: 230-244Crossref PubMed Scopus (43) Google Scholar). Finally, we discuss some recent studies that successfully combine genetic lineage tracing with single-cell transcriptomics, highlighting the power of integrating these two techniques. In the short time span since the first single-cell transcriptome sequencing technique was published in 2009 (Tang et al., 2009Tang F. Barbacioru C. Wang Y. Nordman E. Lee C. Xu N. Wang X. Bodeau J. Tuch B.B. Siddiqui A. et al.mRNA-Seq whole-transcriptome analysis of a single cell.Nat. Methods. 2009; 6: 377-382Crossref PubMed Scopus (787) Google Scholar), an impressive amount of new techniques have become available (Hashimshony et al., 2012Hashimshony T. Wagner F. Sher N. Yanai I. CEL-seq: Single-cell RNA-seq by multiplexed linear amplification.Cell Rep. 2012; 2: 666-673Abstract Full Text Full Text PDF PubMed Scopus (368) Google Scholar, Hashimshony et al., 2016Hashimshony T. Senderovich N. Avital G. Klochendler A. de Leeuw Y. Anavy L. Gennert D. Li S. Livak K.J. Rozenblatt-Rosen O. et al.CEL-Seq2: Sensitive highly-multiplexed single-cell RNA-Seq.Genome Biol. 2016; 17: 77Crossref PubMed Google Scholar, Islam et al., 2014Islam S. Zeisel A. Joost S. La Manno G. Zajac P. Kasper M. Lönnerberg P. Linnarsson S. Quantitative single-cell RNA-seq with unique molecular identifiers.Nat. Methods. 2014; 11: 163-166Crossref PubMed Scopus (311) Google Scholar, Klein et al., 2015Klein A.M. Mazutis L. Akartuna I. Tallapragada N. Veres A. Li V. Peshkin L. Weitz D.A. Kirschner M.W. Droplet barcoding for single-cell transcriptomics applied to embryonic stem cells.Cell. 2015; 161: 1187-1201Abstract Full Text Full Text PDF PubMed Scopus (538) Google Scholar, Macosko et al., 2015Macosko E.Z. Basu A. Satija R. Nemesh J. Shekhar K. Goldman M. Tirosh I. Bialas A.R. Kamitaki N. Martersteck E.M. et al.Highly parallel genome-wide expression profiling of individual cells using nanoliter droplets.Cell. 2015; 161: 1202-1214Abstract Full Text Full Text PDF PubMed Scopus (920) Google Scholar, Muraro et al., 2016Muraro M.J. Dharmadhikari G. Grün D. Groen N. Dielen T. Jansen E. van Gurp L. Engelse M.A. Carlotti F. de Koning E.J. van Oudenaarden A. A single-cell transcriptome atlas of the human pancreas.Cell Syst. 2016; 3: 385-394Abstract Full Text Full Text PDF PubMed Scopus (462) Google Scholar, Picelli et al., 2013Picelli S. Björklund A.K. Faridani O.R. Sagasser S. Winberg G. Sandberg R. Smart-seq2 for sensitive full-length transcriptome profiling in single cells.Nat. Methods. 2013; 10: 1096-1098Crossref PubMed Scopus (371) Google Scholar). While manual picking and processing of each individual cell was initially required, now thousands to tens of thousands of cells are processed in parallel using microfluidic or robot-based approaches (Klein et al., 2015Klein A.M. Mazutis L. Akartuna I. Tallapragada N. Veres A. Li V. Peshkin L. Weitz D.A. Kirschner M.W. Droplet barcoding for single-cell transcriptomics applied to embryonic stem cells.Cell. 2015; 161: 1187-1201Abstract Full Text Full Text PDF PubMed Scopus (538) Google Scholar, Macosko et al., 2015Macosko E.Z. Basu A. Satija R. Nemesh J. Shekhar K. Goldman M. Tirosh I. Bialas A.R. Kamitaki N. Martersteck E.M. et al.Highly parallel genome-wide expression profiling of individual cells using nanoliter droplets.Cell. 2015; 161: 1202-1214Abstract Full Text Full Text PDF PubMed Scopus (920) Google Scholar, Muraro et al., 2016Muraro M.J. Dharmadhikari G. Grün D. Groen N. Dielen T. Jansen E. van Gurp L. Engelse M.A. Carlotti F. de Koning E.J. van Oudenaarden A. A single-cell transcriptome atlas of the human pancreas.Cell Syst. 2016; 3: 385-394Abstract Full Text Full Text PDF PubMed Scopus (462) Google Scholar). Two recently published techniques completely eliminated the need for single-cell isolation by using a pool and split strategy in which cells are first labeled in multiple groups, followed by an iteration of pooling, splitting, and labeling to ensure each single-cell obtains a unique set of labels (Cao et al., 2017Cao J. Packer J.S. Ramani V. Cusanovich D.A. Huynh C. Daza R. Qiu X. Lee C. Furlan S.N. Steemers F.J. et al.Comprehensive single-cell transcriptional profiling of a multicellular organism.Science. 2017; 357: 661-667Crossref PubMed Scopus (68) Google Scholar, Rosenberg et al., 2018Rosenberg A.B. Roco C.M. Muscat R.A. Kuchina A. Sample P. Yao Z. Gray L. Peeler D.J. Mukherjee S. Chen W. et al.Single-cell profiling of the developing mouse brain and spinal cord with split-pool barcoding.Science. 2018; 360: 176-182Crossref PubMed Scopus (0) Google Scholar). A more detailed overview of the available single-cell mRNA sequencing techniques and their properties can be found in Haque et al., 2017Haque A. Engel J. Teichmann S.A. Lönnberg T. A practical guide to single-cell RNA-sequencing for biomedical research and clinical applications.Genome Med. 2017; 9: 75Crossref PubMed Scopus (0) Google Scholar and Papalexi and Satija, 2018Papalexi E. Satija R. Single-cell RNA sequencing to explore immune cell heterogeneity.Nat. Rev. Immunol. 2018; 18: 35-45Crossref PubMed Scopus (4) Google Scholar. With the massive increase in throughput of single-cell mRNA sequencing techniques, multidimensional data analysis becomes increasingly important. Numerous algorithms have been developed to cluster cells, extract significantly differentially expressed genes between clusters of cells, identify outlier cells, and allow visual representation of datasets in two dimensions. These algorithms along with other important aspects of the design and analysis of single-cell mRNA sequencing experiments will not be discussed in depth but are reviewed in Grün and van Oudenaarden, 2015Grün D. van Oudenaarden A. Design and analysis of single-cell sequencing experiments.Cell. 2015; 163: 799-810Abstract Full Text Full Text PDF PubMed Scopus (114) Google Scholar and Yuan et al., 2017Yuan G.C. Cai L. Elowitz M. Enver T. Fan G. Guo G. Irizarry R. Kharchenko P. Kim J. Orkin S. et al.Challenges and emerging directions in single-cell analysis.Genome Biol. 2017; 18: 84Crossref PubMed Scopus (10) Google Scholar. Single-cell transcriptomics allows one to investigate the transcriptional state of thousands of individual single cells thereby reliably capturing cell-type diversity in heterogeneous samples. When applied to a developing or differentiating biological system, many cells transition between different states. If sufficient amounts of cells in these transition states are captured, differentiation trajectories through which tissues are built or maintained can be accurately predicted. These differentiation trajectories can then be exploited to probe kinship among different cell types and to identify genes essential for transitions along these trajectories. However, the difficulty with single-cell transcriptomics data lies in its inherent noisiness and dropout effects (lowly expressed genes are difficult to detect due to technical limitations). Over the last few years, an impressive amount of computational methods have been developed to place cells on differentiation trajectories. Most of these methods rely on the assumption that cells with similar expression profiles arise from the same lineage and that cells with more similarity between their expression profiles are closely related. Here, we will discuss some of those algorithms and the biological systems they have been applied to. The majority of the differentiation trajectory reconstruction methods rely on some form of dimensionality reduction. One of the first and commonly applied algorithms, Monocle (Trapnell et al., 2014Trapnell C. Cacchiarelli D. Grimsby J. Pokharel P. Li S. Morse M. Lennon N.J. Livak K.J. Mikkelsen T.S. Rinn J.L. The dynamics and regulators of cell fate decisions are revealed by pseudotemporal ordering of single cells.Nat. Biotechnol. 2014; 32: 381-386Crossref PubMed Scopus (459) Google Scholar), uses independent-component analysis (ICA) to project all cells in a low dimensional space (usually 2 dimensions) (Figure 2A). This is followed by Minimum Spanning Tree (MST) construction and the definition of a backbone connecting the most and least differentiated cells. All remaining cells are then projected on the backbone, resulting in a 1-dimensional ordering of all cells. This dimension is termed pseudotime, which represents the predicted lineage trajectory of the studied sample. In Monocle, the pseudotime ordering of the cells is not directional, meaning that high pseudotime could either mean most or least differentiated cells; however, users can define a root cell giving the MST a starting point. Monocle was originally used to reconstruct the differentiation trajectory of developing human skeletal muscle myoblasts. One limitation of Monocle is its inability to allow for bifurcations in the lineage prediction, so only linear differentiation systems can be analyzed. This issue has been resolved in the second edition, Monocle2 (Qiu et al., 2017Qiu X. Mao Q. Tang Y. Wang L. Chawla R. Pliner H.A. Trapnell C. Reversed graph embedding resolves complex single-cell trajectories.Nat. Methods. 2017; 14: 979-982Crossref PubMed Scopus (61) Google Scholar). Monocle2 builds the lineage tree in a higher dimensional space, retaining more data for highly intricate differentiation trajectories. Since their publication, Monocle and Monocle2 have been used in numerous studies to predict differentiation trajectories of many developing or differentiating systems. Among these are studies that unravel the lineage trajectories of several types of neurons (Bardy et al., 2016Bardy C. van den Hurk M. Kakaradov B. Erwin J.A. Jaeger B.N. Hernandez R.V. Eames T. Paucar A.A. Gorris M. Marchand C. et al.Predicting the functional states of human iPSC-derived neurons with single-cell RNA-seq and electrophysiology.Mol. Psychiatry. 2016; 21: 1573-1588Crossref PubMed Scopus (12) Google Scholar, Camp et al., 2015Camp J.G. Badsha F. Florio M. Kanton S. Gerber T. Wilsch-Bräuninger M. Lewitus E. Sykes A. Hevers W. Lancaster M. et al.Human cerebral organoids recapitulate gene expression programs of fetal neocortex development.Proc. Natl. Acad. Sci. USA. 2015; 112: 15672-15677Crossref PubMed Scopus (30) Google Scholar, Dulken et al., 2017Dulken B.W. Leeman D.S. Boutet S.C. Hebestreit K. Brunet A. Single-cell transcriptomic analysis defines heterogeneity and transcriptional dynamics in the adult neural stem cell lineage.Cell Rep. 2017; 18: 777-790Abstract Full Text Full Text PDF PubMed Scopus (18) Google Scholar, Hanchate et al., 2015Hanchate N.K. Kondoh K. Lu Z. Kuang D. Ye X. Qiu X. Pachter L. Trapnell C. Buck L.B. Single-cell transcriptomics reveals receptor transformations during olfactory neurogenesis.Science. 2015; 350: 1251-1255Crossref PubMed Scopus (61) Google Scholar, Lacar et al., 2016Lacar B. Linker S.B. Jaeger B.N. Krishnaswami S. Barron J. Kelder M. Parylak S. Paquola A. Venepally P. Novotny M. et al.Nuclear RNA-seq of single neurons reveals molecular signatures of activation.Nat. Commun. 2016; 7: 11022Crossref PubMed Google Scholar, Lisi et al., 2017Lisi V. Singh B. Giroux M. Guzman E. Painter M.W. Cheng Y.C. Huebner E. Coppola G. Costigan M. Woolf C.J. Kosik K.S. Enhanced neuronal regeneration in the CAST/Ei mouse strain is linked to expression of differentiation markers after injury.Cell Rep. 2017; 20: 1136-1147Abstract Full Text Full Text PDF PubMed Scopus (0) Google Scholar, Marques et al., 2016Marques S. Zeisel A. Codeluppi S. van Bruggen D. Mendanha Falcão A. Xiao L. Li H. Häring M. Hochgerner H. Romanov R.A. et al.Oligodendrocyte heterogeneity in the mouse juvenile and adult central nervous system.Science. 2016; 352: 1326-1329Crossref PubMed Google Scholar, Treutlein et al., 2016Treutlein B. Lee Q.Y. Camp J.G. Mall M. Koh W. Shariati S.A. Sim S. Neff N.F. Skotheim J.M. Wernig M. Quake S.R. Dissecting direct reprogramming from fibroblast to neuron using single-cell RNA-seq.Nature. 2016; 534: 391-395Crossref PubMed Scopus (243) Google Scholar), progenitors and stem cells of the hematopoietic system (Mass et al., 2016Mass E. Ballesteros I. Farlik M. Halbritter F. Günther P. Crozet L. Jacome-Galarza C.E. Händler K. Klughammer J. Kobayashi Y. et al.Specification of tissue-resident macrophages during organogenesis.Science. 2016; 353: 353Crossref PubMed Scopus (128) Google Scholar, Qiu et al., 2017Qiu X. Mao Q. Tang Y. Wang L. Chawla R. Pliner H.A. Trapnell C. Reversed graph embedding resolves complex single-cell trajectories.Nat. Methods. 2017; 14: 979-982Crossref PubMed Scopus (61) Google Scholar, Stubbington et al., 2016Stubbington M.J.T. Lönnberg T. Proserpio V. Clare S. Speak A.O. Dougan G. Teichmann S.A. T cell fate and clonality inference from single-cell transcriptomes.Nat. Methods. 2016; 13: 329-332Crossref PubMed Scopus (78) Google Scholar, Tang et al., 2017Tang Q. Iyer S. Lobbardi R. Moore J.C. Chen H. Lareau C. Hebert C. Shaw M.L. Neftel C. Suva M.L. et al.Dissecting hematopoietic and renal cell heterogeneity in adult zebrafish at single-cell resolution using RNA sequencing.J. Exp. Med. 2017; 214: 2875-2887Crossref PubMed Scopus (1) Google Scholar), the placenta (Tsang et al., 2017Tsang J.C.H. Vong J.S.L. Ji L. Poon L.C.Y. Jiang P. Lui K.O. Ni Y.B. To K.F. Cheng Y.K.Y. Chiu R.W.K. Lo Y.M.D. Integrative single-cell and cell-free plasma RNA transcriptomics elucidates placental cellular dynamics.Proc. Natl. Acad. Sci. USA. 2017; 114: 7786-7795Crossref PubMed Scopus (0) Google Scholar), and hair follicles (Joost et al., 2016Joost S. Zeisel A. Jacob T. Sun X. La Manno G. Lönnerberg P. Linnarsson S. Kasper M. Single-cell transcriptomics reveals that differentiation and spatial signatures shape epidermal and hair follicle heterogeneity.Cell Syst. 2016; 3: 221-237Abstract Full Text Full Text PDF PubMed Google Scholar). Several other algorithms, including SLICE (Guo et al., 2017Guo M. Bao E.L. Wagner M. Whitsett J.A. Xu Y. SLICE: Determining cell differentiation and lineage based on single cell entropy.Nucleic Acids Res. 2017; 45: 54Crossref PubMed Scopus (0) Google Scholar), TSCAN (Ji and Ji, 2016Ji Z. Ji H. TSCAN: Pseudo-time reconstruction and evaluation in single-cell RNA-seq analysis.Nucleic Acids Res. 2016; 44: 117Crossref PubMed Scopus (0) Google Scholar), Waterfall (Shin et al., 2015Shin J. Berg D.A. Zhu Y. Shin J.Y. Song J. Bonaguidi M.A. Enikolopov G. Nauen D.W. Christian K.M. Ming G.L. Song H. Single-cell RNA-seq with waterfall reveals molecular cascades underlying adult neurogenesis.Cell Stem Cell. 2015; 17: 360-372Abstract Full Text Full Text PDF PubMed Scopus (178) Google Scholar), SCUBA (Marco et al., 2014Marco E. Karp R.L. Guo G. Robson P. Hart A.H. Trippa L. Yuan G.C. Bifurcation analysis of single-cell gene expression data reveals epigenetic landscape.Proc. Natl. Acad. Sci. USA. 2014; 111: 5643-5650Crossref PubMed Scopus (0) Google Scholar), and Slingshot (Street et al., 2017Street K. Risso D. Fletcher R.B. Das D. Ngai J. Yosef N. Purdom E. Dudoit S. Slingshot: Cell lineage and pseudotime inference for single-cell transcriptomics.bioRxiv. 2017; (Published online April 19, 2017)https://doi.org/10.1101/128843Crossref Google Scholar), employ a similar strategy as Monocle by first reducing dimensionality of the data through principal-component analysis (PCA), ICA, or t-stochastic neighbor embedding (tSNE) and then constructing a MST or fitting a smooth curve and finally projecting single cells on the pseudotime axis (Figure 2A). MST construction in SLICE, TSCAN, and Waterfall is less sensitive to outliers since these algorithms use predefined clusters of cells rather than single cells. SLICE uses transcriptome entropy as a measure for differentiation in order to identify the least differentiated cells in the population and thereby creates a pseudotime starting point. The application of TSCAN to a single-cell dataset containing primitive hematopoiesis cells revealed the importance of the HOPX transcription factor for blood formation (Palpant et al., 2017Palpant N.J. Wang Y. Hadland B. Zaunbrecher R.J. Redd M. Jones D. Pabon L. Jain R. Epstein J. Ruzzo W.L. et al.Chromatin and transcriptional analysis of mesoderm progenitor cells identifies HOPX as a regulator of primitive hematopoiesis.Cell Rep. 2017; 20: 1597-1608Abstract Full Text Full Text PDF PubMed Google Scholar). SCUBA reduces dimensionality of the data using tSNE followed by the fitting of a smooth curve. SCUBA has been used to identify a population of cells in transition between Lgr5+ stem cells and more mature cells in the mouse small intestine. On top of that, SCUBA identified a set of key genes that change expression during this transition (Kim et al., 2016Kim T.H. Saadatpour A. Guo G. Saxena M. Cavazza A. Desai N. Jadhav U. Jiang L. Rivera M.N. Orkin S.H. et al.Single-cell transcript profiles reveal multilineage priming in early progenitors derived from Lgr5(+) intestinal stem cells.Cell Rep. 2016; 16: 2053-2060Abstract Full Text Full Text PDF PubMed Scopus (18) Google Scholar). In contrast to SLICE, SCUBA, and TSCAN, Slingshot takes any form of dimensional reduction, constructs a MST, and then further refines this tree by fitting smooth curves through all of the major MST branches. Single cells are then projected onto their closest curve resulting in ordered lineage trajectories with bifurcations. Slingshot has recently been used to predict the cell-fate potentials and branch points in the lineage trajectories of olfactory stem cells (Fletcher et al., 2017Fletcher R.B. Das D. Gadye L. Street K.N. Baudhuin A. Wagner A. Cole M.B. Flores Q. Choi Y.G. Yosef N. et al.Deconstructing olfactory stem cell trajectories at single-cell resolution.Cell Stem Cell. 2017; 20: 817-830Abstract Full Text Full Text PDF PubMed Scopus (18) Google Scholar). Another class of differentiation trajectory reconstruction algorithms is based on k-nearest neighbor graphs (k-NNGs). In k-NNGs, each cell is connected to its k nearest neighbors, thereby linking similar cells to each other. The first algorithm using k-NNGs was Wanderlust (Bendall et al., 2014Bendall S.C. Davis K.L. Amir A.D. Tadmor M.D. Simonds E.F. Chen T.J. Shenfeld D.K. Nolan G.P. Pe’er D. Single-cell trajectory detection uncovers progression and regulatory coordination in human B cell development.Cell. 2014; 157: 714-725Abstract Full Text Full Text PDF PubMed Scopus (264) Google Scholar), which represents cells as nodes in a collection of k-NNGs, each comprising a subset of the total population of cells (Figure 2B). A user-defined root cell is used to generate a collection of shortest walks from the root cell to all the other cells in each of the graphs. This process results in numerous possible differentiation trajectories, which are then averaged to select the most probable one. Similar to Monocle, Wanderlust only enables the study of linear trajectories, while its successor, Wishbone (Setty et al., 2016Setty M. Tadmor M.D. Reich-Zeliger S. Angel O. Salame T.M. Kathail P. Choi K. Bendall S. Friedman N. Pe’er D. Wishbone identifies bifurcating developmental trajectories from single-cell data.Nat. Biotechnol. 2016; 34: 637-645Crossref PubMed Scopus (122) Google Scholar) allows bifurcations, expanding the repertoire of complex differentiation trajectories that can be studied (Figure 2B). Both Wanderlust and Wishbone were originally designed for Cytometry by Time Of Flight (CyTOF) data, but Wishbone has been adapted so it can be used with single-cell transcriptomics data. Wanderlust and Wishbone were developed to order cells along a developmental axis; however, there are other uses for k-NNGs in single-cell transcriptomics. Markov Affinity-based Graph Imputation of Cells (MAGIC), for instance, locally diffuses gene-expression values in the NNG, thereby smoothening gene expression across highly similar cells (van Dijk et al., 2017van Dijk D. Nainys J. Sharma R. Kathail P. Carr A.J. Moon K.R. Mazutis L. Wolf G. Krishnaswamy S. Pe’er D. MAGIC: A diffusion-based imputation method reveals gene-gene interactions in single-cell RNA-sequencing data.bioRxiv. 2017; (Published online February 25, 2017)https://doi.org/10.1101/111591Crossref Scopus (0) Google Scholar). This reduces the dropout effects often observed for lowly expressed genes. Other graph-based algorithms include approximate graph abstraction (AGA) (Wolf et al., 2017Wolf F.A. Hamey F. Plass M. Solana J. Dahlin J.S. Gottgens B. Rajewsky N. Simon L. Theis F.J. Graph abstraction reconciles clustering with trajectory inference through a topology preserving map of single cells.bioRxiv. 2017; (Published online October 25, 2017)https://doi.org/10.1101/208819Crossref Google Scholar) and population balance analysis (PBA) (Weinreb et al., 2018Weinreb C. Wolock S. Tusi B.K. Socolovsky M. Klein A.M. Fundamental limits on dynamic inference from single-cell snapshots.Proc. Natl. Acad. Sci. USA. 2018; 115: 2467-2476Crossref PubMed Scopus (1) Google Scholar). AGA averages single cells on the NNG into clusters before constructing the differentiation tree and should therefore be less sensitive to outl
Referência(s)