Inference of Protein Assembly in Crystals
A repository for protein quaternary structures inferred from crystals using classification and symmetry

Back to IPACdb
Searching the Repository
Interpreting Result
Understanding IPACdb Detail

Searching the Repository
The search parameters of IPACdb server has been classified into three categories (marked by three different background colors) - Protein Information, Crystallographic Information, and Assembly Information (Figure 1).

Figure 1: The search page of IPACdb server.

Protein Information offers search related to data like PDB ID, SCOP ID, UniProtKB ID, title, name of molecule and from which source the protein has been derived. The repository contains only those proteins whose structure has been solved using X-Ray crystallographic method. Thus, the experimental method related search parameters like resolution, R-factor, and space group information is provided under Crystallographic Information category. Assembly Information category provides search related to protein quaternary structure as modeled by Mitra & Pal (IPAC).

Let us assume, we want to search the database for those homomeric proteins which are tetramer and source is homo sapiens. In order to restrict to valid proteins only (and not peptides), we will select proteins with the subunit length >25 and the structure solved at resolution and R-factor within 2.5 Å and 0.2, respectively. At the same time, we do not want very small complexes and thus will restrict the ASA and BSA with a lower cut-off value of 1000.0 Å2 and 300.0 Å2, respectively. The selected parameters for this search is shown in Figure 1. A result page with all those proteins satisfying the searching criteria is displayed upon pressing the Submit button.

Please note that all the parameters are considered in conjunction to search the repository.

Interpreting Result
The result page will list the summary information of all those proteins which satisfy user mentioned criteria. A toy example page showing 12 proteins selected is shown in Figure 2.

Figure 2: An example Result Page from IPACdb server. Each row corresponds to a complex. The list is sorted by the PDB ID.

First column, which indicates PDB ID is linked to RCSB website, is to provide further information from Protein Data Bank.

Accessible surface area (ASA) and buried surface area (BSA) of the modeled quaternary structure is given in second and third column, respectively. Note that BSA is the sum of the amount of ASA of each atom from all the subunits which lost their ASA due to complex formation. Thus, monomers (absent in this example) do not have any BSA.

The fourth column indicating the quaternary state is linked to the coordinate of the IPAC modeled structure. User can download this coordinate file which is in PDB format.

Next column is the information of the point group symmetry, as seen in the modeled structure. The subunits with different axis of symmetry has been grouped and their symmetry is separated by a comma (,).

The penultimate column contains the Chains-of-assembly information. That means, which chains and at what number are participating in the formation of protein assembly. The integer attached with a literal (chain identifier) shows the number of chains with that chain id involved in the formation of quaternary structure.

Finally, to facilitate the easy online visualization of the structure a JMol button is provided at the last column. The button will help to display the structure in the browser (Figure 4). The only requirement is your browser must have the latest version of java enabled.

IPACdb is being updated periodically. In case IPACdb does not contain your protein of interest, please inform us to inclue that in our next update. Our next update is scheduled on November, 2012.

The penultimate (Assembly) column is linked to the IPACdb detail page. Upon mouse click, the detail information corresponding to each protein will be displayed in a new page. For example, upon mouse click on "C-D-A-B" (at row 11 (PDB ID: 1XU9) and column 6 (Assembly)) of Figure 2, a detail page corresponding to PDB ID: 1XU9 will be opened (Figure 3).

Understanding IPACdb Detail
IPACdb detail page (Figure 3) contains information regarding protein structure as extracted from RCSB, UniProt, SCOP web servers and IPAC method.

Title of the PDB entry, name of the molecule and source of the molecule along with the chain id and length are extracted from RCSB website and is grouped under "Protein Information". At this point the proteins are classified into three categories single chain, homomeric complex and heteromeric complex based on the fact whether PDB contains a single chain, multiple identical chains or multiple distinct chains, respectively.

SCOP class and UniProtKB ID has been extracted from SCOP and UniProt repository, respectively. The SCOP ID is included to know the fold class and their presence in different quaternary states. SCOP ID is placed in a square bracket following the chain ids' of the subunits. If a group of chain ids' have different fold and thus SCOP ID, they are separated by semicolon (;) within the square bracket. Different groups of chain id(s) are separated by front slash (/). To ease the access of further information , the UniProtKB and SCOP ID is directly linked to the corresponding uniprot and fold class information respectively.

Figure 3: The detail information on PDB ID: 1XU9, as provided by the IPACdb database server.

Second section is for experimental information. Note that, IPACdb contains only those proteins whose structures are solved by X-ray crystallography. Thus, experimental method is X-Ray crystallography and the section contains resolution, R-factor, space group and unit cell constants. In order to facilitate the understanding of the space group, each space group is linked to the information provided by the Department of Crystallography, Birkbeck College, University of London.

Third section contains a summary of the protein quaternary structure as modeled by IPAC. Only the cyclic point group symmetry is mentioned, as dyad symmetry can be represented by cyclic elements. Chain-of-assembly information is provided to indicate which chains and how many are taking part in the formation of quaternary state. The integer value attached with a chain id indicates the number of that particular chain involved in quaternary structure. A summary on the accessibility (in Å2 unit) of the protein is also maintained at this section. The total accessible surface area (ASA), amount of buried surface area (BSA) and the percentage of buried surface area have been tabulated here. ASA has been computed using NACCESS program. Another optional information at this section is the number of disulphide bonds present in the complex (both, intra- and inter-chain). Absence of this optional information indicates that there does not exist any disulphide bonds in the structure.

The detail analysis of the biological contacts, as identified by the IPAC classifier is described in fourth section. There are ten features which are used by the classifier to test whether the contact is a valid biological contact or not. To have the details of the parameters, including computation method and unit, please follow the link provided at the footnote of the table. Only those contacts are listed which are biological, as per the classifier.

On the right pane of the page, the link to download the modeled structure and the button for Jmol visualization (in web browser) is provided just below the snap shot of the protein quaternary structure. The only requirement for visualization using Jmol is an updated Java version. Figure 4 shows the visualization of PDB ID: 1XU9 in web browser using Jmol.

Finally, to ease the access of more information regarding the selected protein, a table containing the links to sixteen other useful databases is given at the bottom of the right pane. The name of the database is directly linked to the particular protein of that database.

Please feel free to contact us with your queries and comments.

Figure 4: Jmol visualization in web browser. The depicted protein is a tetramer (PDB ID: 1XU9).