proovframe

proovframe corrects frameshift errors in long-read sequencing contigs and raw reads using protein-to-nucleotide alignments against reference databases to restore reading-frame fidelity and improve gene prediction in complex samples such as microbial communities.

Key Features:

• Frameshift correction: Identifies and corrects frameshifts caused by small insertions and deletions (indels) in long-read sequencing data.
• Protein-to-nucleotide alignment: Uses protein-to-nucleotide alignments against reference databases to detect coding-region frameshifts.
• Input types: Applies corrections to assembled contigs and raw long-read sequences.
• Precise indel edits: Performs targeted insertion or deletion of 1-2 bases within aligned regions to restore the reading frame.
• Low-depth applicability: Improves gene prediction accuracy without requiring high sequencing depth, enabling analysis of rare community members.
• Remote homolog compatibility: Effective with remote protein homologs from reference databases.
• Comparable polishing: Demonstrates efficacy comparable to short-read-based polishing methods on assembled data.
• Protein-guided correction: Leverages protein-guided frameshift correction to enhance analyzability of long-read genomic data.

Scientific Applications:

• Gene prediction: Enhances accuracy of gene prediction in long-read assemblies and raw reads by restoring reading frames.
• Metagenomics: Facilitates analysis of complex microbial communities, including rare taxa, by correcting frameshifts in long-read data.
• Assembly polishing alternative: Serves as an alternative or complement to short-read-based polishing for assembled data.
• Raw-read correction: Enables direct correction of raw long reads to support downstream genomic analyses.

Methodology:

proovframe aligns proteins to nucleotide sequences using reference databases to identify frameshift errors and corrects them by inserting or deleting 1-2 bases within the aligned regions of contigs and raw long reads.