hmmscan (genouest)

hmmscan (genouest) scans protein sequences and profile hidden Markov models (profile HMMs) against HMM databases to identify homologous proteins, conserved domains, and sequence-level similarity.

Key Features:

• HMMER3 speed: Implements HMMER3 with an approximately 100-fold speed increase over earlier versions for protein sequence database searches.
• Search modes: Supports searching single protein sequences, multiple sequence alignments or profile HMMs against sequence databases, and protein sequences against the Pfam database.
• Algorithm: Uses probabilistic methods based on profile hidden Markov models for sequence similarity detection.
• Programmatic access: Exposes search methods as RESTful web services for integration into scripted workflows.
• Output formats: Produces tabular results and graphical summaries of search outcomes.

Scientific Applications:

• Protein family and domain identification: Detects membership of sequences in known protein families and conserved domains, including Pfam entries.
• Functional annotation: Infers protein function by sequence similarity to characterized homologs and conserved motifs.
• Comparative genomics: Enables detection of homologs across genomes to inform evolutionary and comparative analyses.
• Discovery of conserved motifs: Identifies conserved sequence motifs that can indicate novel or conserved protein functions.

Methodology:

Performs probabilistic sequence similarity searches using profile hidden Markov models implemented in HMMER3 and supports searches of sequences and profile HMMs against sequence databases including Pfam.