PreDisorder

PreDisorder predicts intrinsically disordered regions in protein sequences using ab initio and consensus prediction methods to support protein-structure and function analyses.

Key Features:

• Ab Initio Prediction (MULTICOM-CMFR): Employs MULTICOM-CMFR to predict disordered regions from sequence without relying on homologous sequence information.
• Meta Prediction (MULTICOM): Incorporates the MULTICOM consensus method that aggregates predictions from multiple sources to improve accuracy and reliability.
• Benchmarking and Performance: Systematically benchmarked against 26 other protein disorder predictors using the CASP8 dataset, demonstrating competitive performance with top ab initio methods and comparability to leading meta and clustering approaches.
• Genomic-scale Prediction: Enables high-throughput, genome-scale prediction suitable for large-scale proteomics and genome-wide analyses of protein disorder.

Scientific Applications:

• Protein Structure Prediction: Identifies intrinsically disordered regions to inform and refine computational models of protein structure.
• Structure Determination: Indicates regions likely to be flexible or recalcitrant to crystallization, guiding experimental design and interpretation.
• Function Annotation: Reveals disordered segments that may contain functional motifs or interaction sites relevant to protein–protein interactions, signaling, and regulatory mechanisms.

Methodology:

Implements an ab initio predictor (MULTICOM-CMFR) and a meta-predictor (MULTICOM); performance was assessed by benchmarking against 26 disorder predictors using the CASP8 dataset.