Pratt (EBI)

Pratt (EBI) identifies conserved patterns in unaligned protein sequences by detecting ambiguous positions and variable-length wildcards and ranking motifs using a nonstatistical significance measure.

Key Features:

• Pattern discovery: Identifies conserved patterns within unaligned protein sequences without requiring alignment.
• Ambiguity handling: Detects patterns that include ambiguous positions.
• Wildcard regions: Accommodates variable-length wildcard (gap) regions within patterns.
• Parameterizable pattern classes: Allows specification of pattern classes via parameters such as degree of ambiguity and the length and number of gaps.
• Scoring: Ranks candidate patterns using a novel nonstatistical significance measure tailored for flexible patterns.
• Motif recovery: Can recover known motifs from PROSITE families and has been applied to recently described protein families.
• Integration: Integrated into the EMBL-EBI Job Dispatcher sequence analysis tools framework and accessible via RESTful and SOAP-based APIs for use with resources like UniProt, InterPro, ENA, and Ensembl Genomes.

Scientific Applications:

• Motif discovery: Discovery of conserved sequence motifs in unaligned protein datasets.
• Motif validation: Recovery and validation of known PROSITE family motifs.
• Protein family analysis: Characterization of recently described and diverse protein families through conserved-pattern identification.
• Integration into workflows: Deployment within EMBL-EBI sequence-analysis workflows and resource pipelines (UniProt, InterPro, ENA, Ensembl Genomes).

Methodology:

Searches unaligned protein sequences for generalized patterns allowing ambiguous positions and variable-length wildcards, defines pattern classes via user-set parameters (degree of ambiguity; length and number of gaps), and ranks patterns using a novel nonstatistical significance measure.