FRAMA: from RNA-seq data to annotated mRNA assemblies.

Area of Science:

• Genomics
• Bioinformatics
• Molecular Biology

Background:

• Second-generation RNA sequencing enables transcriptome characterization but de novo assembly of full-length mRNAs remains challenging.
• Eukaryotic transcriptomes are complex, featuring highly similar paralogs and alternative splicing variants, complicating accurate reconstruction.
• Existing methods struggle with post-assembly tasks like redundancy reduction and transcript refinement.

Purpose of the Study:

• To present FRAMA, a genome-independent tool for de novo mRNA assembly and annotation.
• To address post-assembly challenges including contig redundancy, ortholog assignment, and transcript correction/scaffolding.
• To enable accurate identification of coding sequences from assembled transcripts.

Main Methods:

• Application of FRAMA to de novo transcriptome assembly using Trinity.
• Genome-independent annotation of naked mole-rat mRNAs.
• Scaffolding of fragmented transcripts and correction of misassembled genes.
• Ortholog assignment and coding sequence identification.

Main Results:

• FRAMA annotated 21,984 naked mole-rat mRNAs, including 12,100 full-length CDSs, corresponding to 16,887 genes.
• Scaffolding increased median sequence information by 1.27-fold, and 4774 misassembled genes were detected and corrected.
• FRAMA's gene models demonstrated superior RNA-seq data support compared to other transcript sets and showed competitiveness with genome-based approaches.

Conclusions:

• FRAMA facilitates the de novo construction of low-redundant eukaryotic transcript catalogs.
• The tool enables transcript extension and refinement, providing a basis for downstream analyses.
• FRAMA is publicly available, supporting gene expression studies and comparative transcriptomics.