EffectiveT3

EffectiveT3 predicts putative Type-III secreted proteins (TTSS effectors) by analyzing N-terminal sequence features to identify conserved Type III secretion signals.

Key Features:

• Machine-Learning Approach: Employs a machine-learning model trained on systematic analysis of amino acid composition and secondary structure of N-termini from 100 experimentally verified effector proteins, using features such as amino acid frequencies, short peptides, and residues with specific physico-chemical properties.
• N-terminal Sequence Analysis: Detects a conserved Type III secretion signal within protein N-termini that is taxonomically universal across animal pathogens and plant symbionts.
• High Sensitivity and Selectivity: Achieves approximately 71% sensitivity and 85% selectivity in predicting TTSS effectors.
• Universal Application: Can detect effector proteins even when the respective taxonomic group is excluded from training, indicating broad applicability across bacterial taxa.
• Genome-Wide Analysis Capability: When applied to 739 complete bacterial and archaeal genomes, identifies between 0% and 12% putative TTSS effectors per genome.

Scientific Applications:

• Effector Discovery: Identification of novel TTSS effector proteins from genomic and proteomic sequences.
• Secretion Mechanism Studies: Characterization of type III secretion mechanisms and the roles of effectors in pathogen-host and symbiont-host interactions.
• Evolutionary Analysis: Investigation of evolutionary processes shaping N-terminal secretion signals, noting absence of clear fusion-based acquisition patterns and supporting convergent evolutionary scenarios.

Methodology:

Analyzes N-terminal sequence features using a machine-learning model developed from systematic analysis of amino acid composition and secondary structure of N-termini from 100 experimentally verified effector proteins, focusing on amino acid frequencies, short peptides, and residues with specific physico-chemical properties.