ML-locMLFE

ML-locMLFE predicts multi-label protein subcellular localization (SCL), determining specific protein locations within cellular environments to support functional analysis and clinical research on pathogen-associated diseases such as COVID-19.

Key Features:

• Multi-Information Fusion: Integrates six feature extraction methods: pseudo amino acid composition; encoding based on grouped weight; gene ontology; multi-scale continuous and discontinuous features; residue probing transformation; and evolutionary distance transformation.
• Dimensionality Reduction: Employs a multi-label information latent semantic index method to reduce redundancy and emphasize essential features.
• Feature-Induced Labeling Information Enrichment: Incorporates multi-label learning with feature-induced labeling information enrichment to refine SCL predictions.

Scientific Applications:

• Functional and host-pathogen studies: Investigating protein function and behavior across organisms, including viral pathogens.
• Clinical research: Informing analyses relevant to clinical treatments, including studies related to COVID-19.
• Taxonomic breadth: Supporting research on Gram-positive and Gram-negative bacteria, plant proteins, and viral proteins.

Methodology:

Feature extraction using six specified methods; multi-label information latent semantic indexing for dimensionality reduction; and multi-label learning with feature-induced labeling information enrichment for SCL prediction.