1.      Tutorial

The tutorial includes two parts: 1) search with PDB structure and 2) search with PDB ID

1.1 Search with PDB structures

1)      Download testing protein structure file (PDB ID: 1P9G) from here.

2)      Click Browse button to select the downloaded PDB file and then click Upload button to upload the PDB file to IPSA server. The email address is optional. If you provide it, the IPSA will send email to notify you when protein substructure comparison and search is done. Three publicly accessible email accounts {ipsatest1, ipsatest2, ipsatest3}@gmail.com  (Password: ipsatest2010) are created for your convenience.


3)      Click Chain ID button to submit.


4)      The IPSA returns Session ID and estimated running time of the submission. You would write down the Session ID or bookmark the link of Session ID and then query the result later.


5)      Query the IPSA results with the Session ID.

IPSA has three modes to show the substructure comparison: M1: top 10 SCOP folds with the best matched protein structure, M2: top 100 matched protein structures from different SCOP folds, and M3: top 10 SCOP folds with all matched proteins from the same fold.

Select one view mode, input Session ID, set threshold of substructure size (default 3), and click Query button.


6)      The IPSA shows the output of substructure comparison of view mode M3 in the following figure. The other view modes have similar output webpage.

The top left panel shows a superimposed view of a query protein chain and a selected result chain from the ranked list in the top right panel. The backbone of query and result protein is represented by while and violet, respectively. Users can check Result Protein and/or Query Protein to view the retrieved chain and/or the query chain. By clicking on a hyperlink below the top left panel, users can see different display themes. The lower panel shows the sequence alignment and statistics of substructure comparison (e.g., number of aligned residues and RMSD). By clicking on residue checkbox below a pair of residue, users can see highlighted residues in the 3D superimposition view. To show one specific substructure, users can first click Clear Display to hide all the substructures and then click substructure bar to show the designed substructure. A sample output can be found here.


1.2 Search with PDB ID

1) Click Search with PDB ID.


2) Select view modes, input PDB ID, set substructure size threshold (default 3) and click Query button.



3) The output webpage is same as Search with PDB structure.

 Any problem, please send email to Admin.


[1] Pin-Hao Chi, Bin Pang, Dmitry Korkin, and Chi-Ren Shyu. Efficient SCOP fold classification and retrieval using index-based protein substructure alignments, Bioinformatics, 2009 25(19):2559-2565. [PDF]