UCL Servers

UCL Servers provide a suite of bioinformatics servers for protein structure prediction and function annotation to support structural biology, genomics, and proteomics research.

Key Features:

• PSIPRED: Predicts protein secondary structure, transmembrane topology, and supports protein fold recognition.
• GenTHREADER (pGenTHREADER): Performs genomic protein fold recognition by threading sequences onto known structures.
• MEMSAT2: Predicts transmembrane protein topology and the orientation and positions of transmembrane helices.
• DISOPRED (DISOPRED2): Predicts intrinsically disordered regions and dynamic disorder within protein sequences.
• DomPred: Predicts domain boundaries to delineate distinct structural or functional domains in protein sequences.
• GTD (Genomic Threading Database): Provides a database of known protein structures and folds to support genomic threading and fold recognition.
• COPS: Compares and classifies protein structural relationships to identify structural similarities and differences.
• MetSite: Predicts metal-binding sites within proteins for analysis of metalloproteins.
• FFPred: Predicts protein function based on structural information to aid functional annotation.
• BioSerf: Fully automated 3D modeling pipeline that employs a fragment-assembly approach to generate de novo protein models and was ranked among the top five servers in CASP8.

Scientific Applications:

• Protein structure prediction: Generation of secondary and tertiary structure models for sequence-based structural inference.
• Transmembrane protein analysis: Topology and helix orientation prediction for membrane protein structural studies.
• Fold recognition and classification: Identification of likely protein folds and comparison to known structures using threading and classification methods.
• Disorder and dynamics analysis: Detection of intrinsically disordered regions relevant to protein flexibility and function.
• Domain delineation: Prediction of domain boundaries to support structural and functional segmentation of proteins.
• Metalloprotein site prediction: Identification of metal-binding residues and sites for studies of metalloprotein function.
• Function annotation from structure: Assignment of likely protein functions based on predicted or known structural features.
• Genome- and proteome-scale annotation: Application of threading, database comparison, and prediction methods for large-scale structural annotation in genomics and proteomics.

Methodology:

Methods include secondary structure and transmembrane topology prediction (PSIPRED, MEMSAT2), protein fold recognition and genomic threading (GenTHREADER; supported by the GTD database), disorder prediction (DISOPRED), domain boundary prediction (DomPred), structural comparison and classification (COPS), metal-binding site prediction (MetSite), function prediction from structural features (FFPred), and fragment-assembly de novo 3D modeling (BioSerf).