Semantic Assembly and Annotation of Draft RNAseq Transcripts without a Reference Genome

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Next-Generation Sequencing (NGS) has made transcriptome sequencing accessible for various organisms, even without a reference genome.
• Traditional de novo assembly methods rely on overlapping short sequence reads, which can be challenging with low coverage or high error rates common in non-model organisms.

Purpose of the Study:

• To develop a computational workflow for reconstructing and functionally annotating expressed gene transcripts from organisms lacking a reference genome.
• To create a method that is tolerant to low coverage, high error rates, and other challenges in de novo assembly.
• To provide a semantic scaffolding step that complements traditional de novo assembly.

Main Methods:

• The workflow assembles unknown transcriptomes by using nearest homologs from public databases as seeds, rather than relying solely on read overlap or unsupervised clustering.
• It considers distant evolutionary relationships to link protein-coding fragments to gene families across multiple genomes.
• The method is designed as an additional scaffolding step following traditional de novo assembly.

Main Results:

• The developed method successfully reconstructed and annotated the transcriptome of the jellyfish Cyanea capillata.
• The approach demonstrated effectiveness in handling challenges associated with de novo assembly in species without high-quality reference genomes.
• The algorithms are implemented in C for parallel computation on high-performance computers and are available under an open-source license.

Conclusions:

• The proposed computational workflow offers a robust solution for transcriptome assembly and functional annotation in species lacking reference genomes.
• This method has broad applicability for gene expression studies in diverse, non-model organisms.
• The open-source software facilitates wider adoption and advancement in comparative genomics and evolutionary biology.