iQSP

iQSP predicts quorum sensing peptides (QSPs) from peptide sequences to identify candidate QSPs and sequence features relevant to bacterial quorum sensing.

Key Features:

• Support Vector Machine (SVM) model: An SVM trained using 18 informative physicochemical properties (PCPs) achieves reported accuracy up to 93.00% and Matthews correlation coefficient (MCC) of 0.86.
• Interpretable rules (IR-QSP): A random forest model derives interpretable rules from the same 18 PCPs to highlight characteristic properties of QSPs.
• Input format: Accepts peptide sequences in FASTA format.

Scientific Applications:

• QSP identification: High-throughput prediction of potential quorum sensing peptides from sequence data.
• Antivirulence drug discovery: Prioritization of QSP candidates for strategies that target or disrupt quorum sensing pathways in bacterial infections.
• Functional characterization: Investigation of physicochemical features that define QSPs and their roles in bacterial communication and pathogenesis.

Methodology:

An SVM is trained on a dataset represented by 18 informative physicochemical properties (PCPs), and a random forest model is used to extract interpretable rules (IR-QSP) from the same PCPs.