@article{fdi:010079057, title = {{S}ouporcell : robust clustering of single-cell {RNA}-seq data by genotype without reference genotypes}, author = {{H}eaton, {H}. and {T}alman, {A}rthur and {K}nights, {A}. and {I}maz, {M}. and {G}affney, {D}. {J}. and {D}urbin, {R}. and {H}emberg, {M}. and {L}awniczak, {M}. {K}. {N}.}, editor = {}, language = {{ENG}}, abstract = {{S}ouporcell clusters single-cell {RNA}-seq data using genotype information without the use of a genotype reference. {M}ethods to deconvolve single-cell {RNA}-sequencing (sc{RNA}-seq) data are necessary for samples containing a mixture of genotypes, whether they are natural or experimentally combined. {M}ultiplexing across donors is a popular experimental design that can avoid batch effects, reduce costs and improve doublet detection. {B}y using variants detected in sc{RNA}-seq reads, it is possible to assign cells to their donor of origin and identify cross-genotype doublets that may have highly similar transcriptional profiles, precluding detection by transcriptional profile. {M}ore subtle cross-genotype variant contamination can be used to estimate the amount of ambient {RNA}. {A}mbient {RNA} is caused by cell lysis before droplet partitioning and is an important confounder of sc{RNA}-seq analysis. {H}ere we develop souporcell, a method to cluster cells using the genetic variants detected within the sc{RNA}-seq reads. {W}e show that it achieves high accuracy on genotype clustering, doublet detection and ambient {RNA} estimation, as demonstrated across a range of challenging scenarios.}, keywords = {}, booktitle = {}, journal = {{N}ature {M}ethods}, volume = {17}, numero = {6}, pages = {615--620}, ISSN = {1548-7091}, year = {2020}, DOI = {10.1038/s41592-020-0820-1}, URL = {https://www.documentation.ird.fr/hor/fdi:010079057}, }