GMV

GMV improves the accuracy and consistency of gene start-site predictions in microbial genomes by applying a genome majority vote algorithm to orthologous gene sets.

Key Features:

• Gene Start Consistency: Aligns gene start sites across orthologous genes in related microbial genomes to address discrepancies among annotations.
• Algorithmic Approach: Implements a genome majority vote algorithm to determine the most probable and consistent gene start positions among ortholog sets.
• Error Correction: Corrects hundreds of gene prediction errors across multiple genomes while introducing minimal new errors, with particular effectiveness in groups of five or ten genomes.
• Validation and Accuracy: Validated accuracy using a set of well-characterized Escherichia coli genes as a benchmark.
• Broad Applicability: Applies to publicly available microbial gene maps to resolve start-site inconsistencies across diverse genomic studies.

Scientific Applications:

• Microbial Genomics: Improves gene mapping accuracy in microbial genomes for downstream analyses.
• Comparative Genomics: Refines alignment of orthologous genes to support evolutionary and functional analyses.
• Genome Annotation: Enhances precision of gene annotations across multiple genomes to increase data reliability.

Methodology:

Initial gene predictions are generated with PRODIGAL on all genomes; pan-reciprocal BLAST is used to identify ortholog sets; predicted start sites are compared in a multiple sequence alignment; when discrepancies occur the algorithm selects a consensus start position based on a majority of high-scoring PRODIGAL predictions and adjusts outlier gene predictions accordingly.