3dSS: 3D structural superposition

W128–W132 Nucleic Acids Research, 2006, Vol. 34, Web Server issue doi:10.1093/nar/gkl036 3dSS: 3D structural superposition K. Sumathi1, P. Ananthalakshmi1, M. N. A. Md. Roshan1 and K. Sekar1,2,* 1 Bioinformatics Centre and 2Supercomputer Education and Research Centre, Indian Institute of Science, Bangalore 560 012, India Received October 6, 2005; Revised and Accepted November 28, 2005 ABSTRACT 3dSS is a web-based interactive computing server, primarily designed to aid researchers, to superpose two or several 3D protein structures. In addition, the server can be effectively used to find the invariant and common water molecules present in the superposed homologous protein structures. The molecular visualization tool RASMOL is interfaced with the server to visualize the superposed 3D structures with the water molecules (invariant or common) in the client machine. Furthermore, an option is provided to save the superposed 3D atomic coordinates in the client machine. To perform the above, users need to enter Protein Data Bank (PDB)-id(s) or upload the atomic coordinates in PDB format. This server uses a locally maintained PDB anonymous FTP server that is being updated weekly. This program can be accessed through our Bioinformatics web server at the URL http://cluster.physics.iisc.ernet.in/3dss/ or http://10.188.1.15/3dss/. INTRODUCTION In the post-genome era, the structural and conformational properties of the 3D protein molecules are of considerable interest owing to its importance in various biological processes. Owing to the recent technological advances like high power tunable synchrotron radiation, powerful number crunching computers and due to ambitious structural genomics programs in different parts of the world, there has been a tremendous increase in the number of protein structures in the Protein Data Bank (PDB) (1). Now there are 34 000 3D structures available in this entity. Analysis of the 3D structure of protein molecules is greatly enhanced by understanding the relationship between the individual protein molecules. Furthermore, knowledge of the 3D structural relationship between different protein molecules is a key issue in understanding the structure and function. In order to find the common structural region, one need to lay one molecule over the other by appropriate rotation and translation and this process is termed as superposition of the 3D structures. Several programs are available in the literature (2–9) for this purpose. Most of these programs are stand-alone versions and have their own merits and demerits. Two most recent ones are web-based servers, namely, SSM (8) and SuperPose (9). The program SSM uses the procedure of matching graphs generated using the secondary structural elements followed by the alignment of Ca atoms of the protein molecule. Using one of the programs (9), SuperPose, we experienced problems while trying to superpose multiple structures as well as portions of molecules. In fact, it was difficult to superpose different subunits available in multi-subunit protein structures. In addition, most of the existing programs use only the first model of the ensemble in the case of structures solved using NMR technique and there is no provision for the users to superpose all the models in the ensemble. It is well known that water molecules play a vital role in protein structures, aiding in stabilizing the protein fold and in ligand design (10–14). In addition, investigations on the invariant water molecules in several well studied homologous protein structures shed light on the specific roles of water molecules such as catalytic, structural and functional (15–18). Thus, it is necessary to find the invariant and common water molecules (for definition see below) in homologous protein structures, for which 3D structural superposition step is crucial. But the existing programs do not have provisions for the users to identify the invariant and common water molecules. Hence, we created a unique computing server to superpose the three-dimensional structures and to find the invariant and common water molecules in homologous protein structures. BACKGROUND The water molecules present in two highly similar (the best example is native structure and its mutant structure) or highly *To whom correspondence should be addressed. Tel: +91 080 23601409; Fax: +91 080 23600085; Email: , This work is dedicated to the late professor M. Sundaralingam The authors wish it to be known that, in their opinion, the first two authors should be regarded as joint First Authors  The Author 2006. Published by Oxford University Press. All rights reserved. The online version of this article has been published under an open access model. Users are entitled to use, reproduce, disseminate, or display the open access version of this article for non-commercial purposes provided that: the original authorship is properly and fully attributed; the Journal and Oxford University Press are attributed as the original place of publication with the correct citation details given; if an article is subsequently reproduced or disseminated not in its entirety but only in part or as a derivative work this must be clearly indicated. For commercial re-use, please contact Nucleic Acids Research, 2006, Vol. 34, Web Server issue homologous structures (the inhibitor free and inhibitor bound structure) are known as invariant water molecules. Further, such situation is also possible in multi-subunit protein structures. For example, if a molecule has four identical subunits, the water molecules that interact with the residues in the same position in different subunits (e.g. subunit A and B) can be considered as invariant water molecules. On the other hand, common water molecules are those, which lie at the interface and interact with the selected subunits. In the computing server, two widely recognized programs STAMP (19) and ProFit (A. C. R. Martin, http://www.bioinf. org.uk/software/profit/) are deployed for superposition purposes. The program STAMP uses multiple sequence alignment using the amino acid sequence information followed by an initial superposition of structures. In contrast, the program ProFit uses the McLachlan fitting algorithm, essentially a steepest descent minimization (3). The user-friendly molecular visualization tool RASMOL (20) is interfaced to view the superposed molecules in the client machine. This server is developed using PERL, HTML and JAVASCRIPT. Ploticus [Copy right 1998–2002, Stephan C. Grugg ()], a data display engine is used for generating plots to display root mean square deviation (r.m.s.d.) graphically. DATA PRESENTATION AND AVAILABILITY The software is developed and optimized for Intel based Solaris (Version 10.0) and is driven by 3.0 GHz pentium IV W129 processor equipped with 2 GB RD RAM. This operating system is chosen for better security, scalability and reliability. The software and its functionalities are well tested on Windows 95/98/2000, Linu (...truncated)