Borf

Borf predicts open reading frames (ORFs) in de-novo assembled transcriptomes, accounting for strand-specific RNA-Seq libraries and incomplete transcript assemblies to improve ORF annotation accuracy.

Key Features:

• Strand-Specificity Consideration: Annotates ORFs exclusively on the sense strand for strand-specific RNA-Seq libraries to reduce false-positive predictions from antisense sequences.
• Start Site Selection Accuracy: Classifies sequences upstream of the first start codon as 5' untranslated regions (5' UTRs) in fully assembled transcripts or as ORF sequence in 5' incomplete transcripts to improve start-site annotation.
• Handling Incomplete Transcripts: Determines an optimal upstream-length cutoff to distinguish transcripts that lack an in-frame upstream stop codon as complete versus 5' incomplete.

Scientific Applications:

• RNA-Seq analysis: Reduces false-positive ORF predictions in analyses of strand-specific RNA-Seq data.
• Transcriptome annotation: Improves start-site prediction and ORF annotation in de-novo assembled transcriptomes, including analyses of well-annotated species.

Methodology:

Implemented in Python3 and evaluated using a gold-standard dataset comprising four de-novo transcriptome assemblies of well-annotated species.