ScBiLR:Imputing Dropout Events In Single Cell RNA Sequencing Data Via Bilinear Regression

DNA sequencing technology and next generation sequencing approaches for high-throughput RNA sequencing are experiencing tremendous growth.Single-cell RNA sequencing(scRNAseq)technique is becoming increasingly popular in transcriptome studies.While previous bulk RNAseq measures average gene expression levels across cells by ignoring potential cell-to-cell heterogeneity,scRNAseq provides an unbiased characterization of gene expression at each single-cell level.Single cell RNA sequencin technology enables whole transcriptome profiling at single cell resolution and holds great promises in many biological and medical applications.Nevertheless,scRNA-seq often fails to capture expressed genes,leading to the prominent dropout problem.In addition,unlike bulk RNA sequencing technology that each gene expression value is the average of all cells,single cell data reflects the expression values of genes in all cells.However,the reality is that some genes are not expressed at all in some cells,so the data will be sparse,and due to the dropouts,the data will be sparser,which causes many problems in down-stream analysis.Imputation of these dropout values thus becomes an essential step in scRNA-seq data analysis.In this paper,we model the dropout imputation problem as bilinear regression.Firstly,a similarity matrix with prior information is constructed,which helps select subsets for coefficients of the following bilinear regression.Then,the lasso regression for rows or columns is estimated by iteration,and the estimated value of convergence condition is used as the final imputation value.Some downstream analysis shows that scBiLR can accurately recover the dropout values and help to improve differential expression analysis and clustering analysis.