Augmenting transcriptome assembly combinatorially

TL;DR

This paper introduces a combinatorial approach to improve transcriptome assembly by integrating results from multiple assembly tools, addressing the limitations of individual methods in sensitivity and accuracy.

Contribution

It presents a novel method to augment transcriptome assemblies by combining outputs from various tools, based on a comprehensive comparative analysis.

Findings

No single assembler achieves full sensitivity and specificity.

Combining multiple assemblies improves transcript recovery.

The approach reduces errors and enhances assembly completeness.

Abstract

RNA-seq allows detection and precise quantification of transcripts, provides comprehensive understanding of exon/intron boundaries, aids discovery of alternatively spliced isoforms and fusion transcripts along with measurement of allele-specific expression. Researchers interested in studying and constructing transcriptomes, especially for non-model species, often face the conundrum of choosing from a number of available de novo and genome-guided assemblers. A comprehensive comparative study is required to assess and evaluate their efficiency and sensitivity for transcript assembly, reconstruction and recovery. None of the popular assembly tools in use today achieves requisite sensitivity, specificity or recovery of full-length transcripts on its own. Hence, it is imperative that methods be developed in order to augment assemblies generated from multiple tools, with minimal compounding of error. Here, we present an approach to combinatorially augment transciptome assembly based on a rigorous comparative study of popular de novo and genome-guided transcriptome assembly tools.