PPFINDER

PPFINDER identifies processed pseudogenes in genome annotations (GTF) and refines gene predictions by masking retrotransposed intronless sequences that can be misannotated as genes.

Key Features:

• Dual Methodology for Pseudogene Identification: Employs two distinct methods to detect processed pseudogenes within mammalian gene annotations.
• Integration with Gene Prediction via Iterative Masking: Masks identified pseudogenes and re-runs gene prediction iteratively until predictions converge, preventing pseudogene-derived false positives.
• Freestanding Pseudogene Removal without Known-Gene Libraries: Operates without relying on libraries of known genes to identify and remove processed pseudogenes from gene predictions.
• Applicability to Newly Sequenced Genomes: Functions independently of pre-existing gene annotations, enabling use on genomes with limited prior gene information.
• Perl Implementation and GTF Support: Implemented in Perl and designed to work with genome annotations formatted in GTF (Gene Transfer Format).

Scientific Applications:

• Improving Gene Prediction Accuracy: Reduces biologically irrelevant gene and exon predictions by removing processed pseudogene interference.
• Functional Genomics: Enhances the biological relevance of predicted gene models used in downstream functional analyses.
• Evolutionary Biology and Comparative Genomics: Clarifies genomic landscapes by distinguishing retrotransposed pseudogenes from functional genes for evolutionary studies.
• N-SCAN Preprocessing: Has been applied to preprocess N-SCAN gene predictions by removing pseudogene interference to improve prediction quality.

Methodology:

Implemented in Perl to operate on GTF genome annotations, PPFINDER applies two explicit pseudogene-identification methods, masks identified processed pseudogenes from the genome sequence, and re-runs gene prediction algorithms iteratively until no further changes in predictions occur.