Genome-guided transcript assembly from integrative analysis of RNA sequence data

The identification of full length transcripts entirely from short-read RNA sequencing data (RNA-seq) remains a challenge in genome annotation pipelines. Here we describe an automated pipeline for genome annotation that integrates RNA-seq and gene-boundary data sets, which we call generalized RNA integration tool, or GRIT. By applying GRIT to Drosophila melanogaster short-read RNA-seq, cap analysis of gene expression (CAGE) and poly(A)-site-seq data collected for the modENCODE project, we recover the vast majority of previously annotated transcripts and double the total number of transcripts cataloged. We find that 20% of protein coding genes encode multiple protein-localization signals, and that, in 20 day old adult fly heads, genes with multiple poly-adenylation sites are more common than genes with alternate splicing or alternate promoters. When compared to the most widely used transcript assembly tools, GRIT recovers a larger fraction of annotated transcripts at higher precision. GRIT will enable the automated generation of high-quality genome annotations without necessitating extensive manual annotation.