WEDGE

The software tool 'WEDGE' addresses the challenge of low capture rates of expressed RNAs in single-cell sequencing data, hindering downstream functional genomics analyses. Existing imputation methods for single-cell transcriptome data struggle with very sparse expression matrices. 'WEDGE' introduces a novel algorithm, WEighted Decomposition of Gene Expression, which employs a biased low-rank matrix decomposition technique for imputing gene expression matrices. It effectively recovers expression matrices, preserves cell-wise and gene-wise correlations, and enhances cell clustering, particularly for sparse datasets. The study underscores the algorithm's potency in imputing sparse expression matrix data, offering researchers a valuable tool to extract biological insights from their single-cell RNA sequencing datasets.