Q. What is GWIDD?
A. GWIDD (Genome-WIde Docking Database) is a comprehensive publicly available resource for structural modeling of protein-protein interactions. It provides three-dimensional structures of complexes between proteins known to interact in living organisms.
Q. What software do I need to use GWIDD?
A. GWIDD is a web application, which means all you need is a recent web browser with JavaScript and Java enabled. Internet Explorer 6 or older is not supported at this time. Screen resolution should be 1024x768 or higher for optimal results.
Q. What is GWIDD current content?
A. Currently GWIDD contains 25,559 structures out of which 10,924 are X-ray structures from PDB and the rest are model structures built by docking methodology (presently homology docking, in the future releases also structure alignment and template free docking). GWIDD contains data for 771 organisms spanning the entire universe of life, from viruses to humans.
Q. What is homology docking?
A. The model of a protein complex is built based on a similar protein complex with a known structure.
Q. What is GWIDD primary source of information on interacting pairs of proteins with known sequences?
A. GWIDD imported interaction data from DIP and BIND databases, which contain information on ~126,000 interacting pairs of proteins with known sequences.
Q. What searches can be performed in GWIDD?
A. User can search GWIDD by keywords, organism, sequence, or PDB file for one or both interacting partners.
Q. What is search by keywords (KEYWORD tab)?
A. The search is performed against any keywords in the sequence description from NCBI sequence database.
Q. What is search by organism (KEYWORD tab)?
A. The search is performed against GWIDD organism annotation. The drop-down menu is provided for user convenience showing all organisms currently in GWIDD. The menu is organized by internal GWIDD classification of living organisms based on UNIPROT taxonomy classification scheme. Alternative way of selecting organisms by taxonomy ID is also offered.
Q. What is search by sequence (SEQUENCE tab)?
A. The search is for interactions in which the input sequences participate. User can either input the sequence directly (in FASTA format) or upload plain-text file with the sequence in FASTA format.
Q. What is search by structure (STRUCTURE tab)?
A. User can upload PDB-formatted file from which GWIDD extracts the sequence and performs the search as in the ‘sequence’ search. In case of a multi-chain PDB file, user can indicate a specific chain.
Q. What is “Protein A” and “Protein B” parts of the search screen?
A. If “Protein A” part only is active then the search is performed only for one of the interacting partners. For example, in searches by sequence, the search results contain all interactions for which the input sequence is one of the partners. If both “Protein A” and “Protein B” parts are active then the search is performed for both interacting partners. For example, if user inputs a keyword in “Protein A” and a sequence in “Protein B” then search results contain all interactions of the input sequence with all sequences having descriptions containing the input keyword.
Q. What search results are displayed on screen?
A. The search results are presented as a list of GWIDD IDs for interactions satisfying the search criteria. Each list item can be expanded displaying GWIDD IDs for both interacting partners, their names and NCBI IDs, along with information on the type of the structure (experimental or model), download link for the PDB-formatted structure file and a link to the visualization screen.
Q. Do all interactions currently in GWIDD have structures?
A. No, only 25,559 out of total 126,897 interactions have the 3D structure available in the current GWIDD release. However, in the near future model structures for interactions that currently have no structure will be built by other modeling methodologies (structure alignment and template free docking) and added to future GWIDD releases.
Q. Is there more than one structure available for a given interaction?
A. No, only one structure is available for each given interaction. In the case of multiple X-ray structures, the selected structure is the most recent one. In the case of multiple docking models, the choice of the structure is based on the assessed quality of the model.
Q. What is visualization screen?
A. The visualization screen provides interactive image (initially in space-filled representation colored by chain) of a given interaction. The graphics is based on Jmol technology, which enables user to manipulate the drawing (rotate, translate, zoom, etc), to change representation (ribbon, sticks etc), to highlight different parts of the molecules, etc.
Q. Is GWIDD content updated?
A. Currently GWIDD is undergoing a major update, expanding the dataset to 800,365 PPI imported from the INTACT and BIOGRID, subjected to our modeling pipeline.
Q. Will GWIDD be developed further?
A. Yes, we will develop GWIDD further by including model structures built by other methodologies as soon as proper benchmarking is completed. The methodologies themselves will be further developed to become more adequate to the large-scale modeling of interactions.