DescribePROT

DescribePROT integrates 10 predictive algorithms to provide amino acid-level predictions of protein structure and function across 83 complete proteomes.

Key Features:

• Integration of predictive algorithms: Combines outputs from ten distinct prediction algorithms to produce residue-level annotations.
• Descriptor set: Provides 13 residue-level descriptors, including sequence conservation, position-specific scoring matrices (PSSMs), secondary structure, solvent accessibility, intrinsic disorder, disordered linkers, signal peptides, motifs for receptor flexibility (MoRFs), and predictions of protein–protein, protein–DNA, and protein–RNA interactions.
• Proteome coverage: Annotations span 83 complete proteomes, including key model organisms.
• Scale of data: Contains approximately 1.4 million proteins covering nearly 600 million amino acids.
• Pre-computed predictions: Comprises about 7.8 billion pre-computed residue-level predictions.

Scientific Applications:

• Protein function analysis: Supports residue-level investigations into protein structure–function relationships.
• Therapeutics and disease research: Provides annotation data applicable to projects targeting therapeutics and disease mechanisms.
• Predictor development and benchmarking: Serves as a resource for developing and benchmarking novel predictors of protein sequence descriptors.
• Comparative proteomics: Enables comparative analyses across complete proteomes and model organisms.

Methodology:

Outputs were generated by integrating ten prediction algorithms to compute 13 residue-level descriptors across 83 complete proteomes, producing ~7.8 billion pre-computed residue-level predictions covering ~1.4 million proteins (~600 million amino acids).