Lizhong Chen

CMDdemux improves single-cell demultiplexing accuracy by integrating hashing and mRNA data, outperforming existing methods on both high- and low-quality datasets.

The paper introduces limma and edgeR pipelines for differential transcript usage analysis that account for read-to-transcript ambiguity, improving power and FDR control in small and large datasets.

edgeR v4 adds a continuous negative binomial for fractional counts, bias-corrected quasi-likelihood for small counts, C-based fitting, and new transcript, exon, methylation and fold-change-threshold analyses.

The paper introduces an improved differential transcript expression analysis pipeline using edgeR v4 and Salmon's Gibbs sampling, showing it is faster and more accurate than previous methods.

GLMsim simulates single cell RNA-seq data incorporating library size, biology, and batch effects, enabling benchmarking of integration methods.