Enhancing transcriptome expression quantification through accurate assignment of long RNA sequencing reads with TranSigner

Long-read RNA sequencing captures transcripts at full lengths, but existing methods for transcriptome profiling using long-read data often produce inconsistent transcript identification and quantification results. Here, we introduce TranSigner, a tool designed to provide read-level support for transcripts in a given transcriptome. TranSigner consists of three modules: read alignment to transcripts, computation of read-to-transcript compatibility scores, and a guided expectation-maximization algorithm to assign reads to transcripts and estimate their abundances. Using simulated and experimental data from three well-studied organisms-Homo sapiens, Arabidopsis thaliana, and Mus musculus-we show that TranSigner achieves accurate read assignments and abundance estimates.