Research Projects

Safe-OmniAMP is a scalable computational workflow designed for comprehensive antimicrobial peptide discovery and safety screening. The system integrates bioinformatics pipelines with machine learning models to identify potential antimicrobial peptides while assessing their safety profiles.

The workflow handles large-scale peptide data and employs advanced computational methods to predict antimicrobial activity, toxicity, and other safety-relevant properties, enabling researchers to prioritize the most promising candidates for experimental validation.

Focus areas: Peptide screening, safety assessment, computational workflows, large-scale data processing.

FusionAMP is a machine-learning framework designed for predicting sequence-divergent antimicrobial peptides. The system leverages evolutionary information and sequence features to identify antimicrobial peptides with diverse sequences across different organisms.

The framework addresses the challenge of antimicrobial peptide conservation by incorporating evolutionary repertoire analysis, enabling the prediction of novel antimicrobial sequences that may not share high sequence similarity with known AMPs but retain similar functional properties.

Focus areas: Sequence prediction, evolutionary analysis, machine learning, peptide classification, functional repertoire.