NetNES

NetNES predicts leucine-rich nuclear export signals (NES) in eukaryotic proteins to identify regulators of protein subcellular localization relevant to transcription, cell cycle regulation, differentiation, and disease.

Key Features:

• Prediction target: Identifies leucine-rich nuclear export signals (NES) in eukaryotic proteins.
• Machine learning methods: Employs neural networks and hidden Markov models for prediction.
• Feature representation: Represents signal properties beyond simple consensus patterns, improving over traditional consensus pattern methods.
• Structural considerations: Incorporates residue accessibility and flexibility as factors in prediction.
• Residue composition: Accounts for properties in addition to hydrophobic residues when defining NESs.
• Validation: Validated using recently discovered NESs.

Scientific Applications:

• Protein subcellular localization analysis: Mapping NESs to study nuclear export and subcellular distribution of proteins.
• Transcription regulation studies: Investigating how NES-mediated export influences transcriptional regulators.
• Cell cycle research: Examining NES roles in proteins involved in cell cycle regulation.
• Cell differentiation research: Exploring NES involvement in processes of cell differentiation.
• Disease and cancer studies: Assessing NES-related changes implicated in cancer progression and other diseases.

Methodology:

Uses a machine learning framework combining neural networks and hidden Markov models, incorporating residue accessibility and flexibility and features beyond hydrophobic residues, with validation on recently discovered NESs.