hmmscan (EBI)

hmmscan (EBI) performs protein sequence similarity searches using profile hidden Markov models (profile HMMs) based on the HMMER3 software to identify homologous sequences and annotate proteins against databases such as Pfam.

Key Features:

• Algorithmic basis: Implements profile HMM probabilistic methods provided by the HMMER3 software suite for sensitive sequence similarity detection.
• Speed and efficiency: Achieves approximately 100-fold speed improvements relative to earlier HMMER versions for rapid database searches.
• Input and target types: Accepts single protein sequences, multiple sequence alignments, or profile HMMs as queries and searches against sequence databases including Pfam.
• Programmatic access: Exposes search methods as RESTful web services to enable integration and automation in computational workflows.
• Result reporting: Produces tabular outputs and graphical summaries that scale to large numbers of matches and enable rapid appraisal of search outcomes.

Scientific Applications:

• Homology detection: Identification of homologous protein sequences across databases.
• Protein annotation: Assignment of protein family labels and functional domains using Pfam and other profile HMM collections.
• Evolutionary analysis: Exploration of evolutionary relationships among protein families via profile-based similarity.
• Functional annotation and comparative genomics: Support for high-throughput functional annotation and comparative genomics within larger bioinformatics workflows.

Methodology:

Searches use profile hidden Markov models implemented in HMMER3 to perform probabilistic sequence similarity searches that model the statistical properties of protein families.