LocNES

LocNES predicts classical nuclear export signals (NESs) in proteins that are substrates of the CRM1-mediated export pathway (CRM1/exportin 1/Karyopherin β2), facilitating identification of motifs that direct nuclear-to-cytoplasmic transport.

Key Features:

• Support Vector Machine (SVM) Model: Employs an SVM-based classifier to distinguish NES-containing peptides from background sequence.
• Comprehensive Feature Set: Integrates amino acid sequence, disorder propensity, and the rank of position-specific scoring matrix (PSSM) scores as input features to the predictive model.
• High Sensitivity and Precision: Demonstrates improved sensitivity and precision relative to existing NES prediction methods according to comparative analyses.

Scientific Applications:

• Understanding Protein Localization: Predicts NESs to inform studies of protein nuclear export and subcellular localization.
• Investigating Disease Mechanisms: Supports identification of NESs in CRM1 cargoes relevant to diseases such as cancer and viral infections.
• Functional Annotation of Proteins: Contributes NES annotations for functional characterization of proteins and their interactions.

Methodology:

Scans protein sequences for peptides matching the consensus pattern of classical NESs and assigns probability scores based on an SVM model that analyzes features including amino acid sequence, disorder propensity, and the rank of position-specific scoring matrix scores.