FGENESH

FGENESH predicts gene structures, pseudogenes, and promoters in eukaryotic genomic DNA using Hidden Markov Models (HMM) to support genome annotation and regulatory element mapping.

Key Features:

• Gene Prediction: Fgenesh++ predicts coding regions in eukaryotic genomic DNA, recovering approximately 91% of coding nucleotides with a specificity of about 90%.
• Pseudogene Identification: Includes an automatic pseudogene finder (PSF) that identifies approximately 90% of manually annotated pseudogenes and can discover novel pseudogenes.
• Promoter Prediction: Incorporates a promoter prediction component (Fprom) that identifies transcription start sites, detecting TATA-box promoters at ~1 false positive per 2,000 bp and TATA-less promoters at ~1 false positive per 650 bp.

Scientific Applications:

• Genome Annotation: Automatic annotation of eukaryotic genomes by predicting gene structures, pseudogenes, and promoter regions.
• Comparative Genomics: Provides predicted gene and regulatory element annotations to support evolutionary and comparative analyses.
• Functional Genomics: Maps transcription start sites to aid investigation of gene regulation and expression patterns.

Methodology:

FGENESH applies Hidden Markov Models (HMM) to analyze genomic sequences and predict multiple genes and their associated features.