Improved transcriptome assembly using a hybrid of long and short reads with StringTie

Abstract

Short-read RNA sequencing and long-read RNA sequencing each have their strengths and weaknesses for transcriptome assembly. While short reads are highly accurate, they are rarely able to span multiple exons. Long-read technology can capture full-length transcripts, but its relatively high error rate often leads to mis-identified splice sites. Here we present a new release of StringTie that performs hybrid-read assembly. By taking advantage of the strengths of both long and short reads, hybrid-read assembly with StringTie is more accurate than long-read only or short-read only assembly, and on some datasets it can more than double the number of correctly assembled transcripts, while obtaining substantially higher precision than the long-read data assembly alone. Here we demonstrate the improved accuracy on simulated data and real data from Arabidopsis thaliana, Mus musculus, and human. We also show that hybrid-read assembly is more accurate than correcting long reads prior to assembly while also being substantially faster. StringTie is freely available as open source software at https://github.com/gpertea/stringtie.

Author summary

Identifying the genes that are active in a cell is a critical step in studying cell development, disease, the response to infection, the effects of mutations, and much more. During the last decade, high-throughput RNA-sequencing data have proven essential in characterizing the set of genes expressed in different cell types and conditions, which has driven a strong need for highly efficient, scalable and accurate computational methods to process these data. As sequencing costs have dropped, ever-larger experiments have been designed, often capturing hundreds of millions or even billions of reads in a single study. These enormous data sets require highly efficient and accurate computational methods for analysis, and they also present opportunities for discovery. Recently developed long-read technology now allows researchers to capture entire transcripts in a single long read, enabling more accurate reconstruction of the full exon-intron structure of genes, although these reads have higher error rates and higher costs. In this study we use the high accuracy of short reads to correct the alignments of long RNA reads, with the goal of improving the identification of novel gene isoforms, and ultimately our understanding of transcriptome complexity.