Kalidas Y

Ph.D. Work

Title - A Novel Algorithmic Suite for Detection, Large Scale Comparison and Design of Ligand Binding Sites in Protein Structures. Molecular recognition in a living cell, happens primarily through protein-ligand interactions at ligand binding sites and its understanding is crucial for studying systems-behaviour and drug side-effects. Ligand binding sites in protein structures are hot-spots of such interactions. My PhD work involved development of algorithms that are highly useful for analysis of ligand binding sites in protein structures from various perspectives and facilitate large-scale studies.

Research supervisors - Dr. Nagasuma Chandra and Prof. S. Raha

Core algorithms developed

• Detection of ligand binding sites (PocketDepth)
• Shape descriptor and fast comparison of ligand binding sites (PocketMatch)
• Structural alignment of binding sites (PocketAlign)
• Structural alignment of ligands and their corresponding binding sites (LigandAlign)
• Design of ligand binding sites for a given ligand (PocketDesign)
• Framework for large-scale binding site comparison and annotation (PocketAnnotate)

Work also involved

• Creation of MPI parallel versions of programs for cluster computers;
• Creation of web-services for enabling public access to the softwares
• Development of wrapper and frame-works primarily in BASH, PERL, AWK and C languages

Research Skills

• Problem analysis and algorithm design
• Problem representation and presentation for discussions
• Computational representation of molecular structures
• Comparison and mining of molecular structures
• Large-scale ‘pocketome’ level comparisons
• Genetic algorithms and numerical methods
• Creation of frame-works and parallel processing modules
• Design and set up of web-services

Courses taken during PhD

• Data structures and algorithms
• Data mining
• Modelling and Simulation
• Computational approaches to drug discovery
• An Introduction to Bioinformatics
• Conformational analysis of biopolymers

Concepts learnt during B. Tech

• Computer organization
• Lex/Yacc compiler tools
• Object-oriented systems
• Computer networks (TCP/IP)
• Linux environment
• Project implemented in B.Tech – implementation of parallel constructs in C language

International interaction

• Presented posters on research work at the ICSB-2008 and ISMB-2009 conferences in Sweden and carried out scientific discussions.
• Visited Chalmers University, Gothenburg, Sweden for one month on a collaborative project on classification of MHC molecules and published the work.
• Visited Leiden University, Amsterdam and delivered seminar in 2009 in Computational biology group.
• Presented work at international conferences organized by IISc

Software skills

• C language, Linux environment, MPI parallel-programming (IBM Bluegene)
• PERL, BASH and AWK scripting
• JAVA (Sun certified), Worked on JCA Adapters on BEA Weblogic server; C++; MATLAB
• Web-server technology (Apache, CGI, JavaScript, AJAX)
• Bioinformatics Software {sequence, structure and phylogeny software}; and Rasmol, Pymol, MOE, InsightII (Accelrys), Biosuite, Phylip, Phylodraw, Cytoscape among others.
• Language tools: Parsing – Lex/Yacc; Documentation – LaTeX, Doxygen

Academic highlights

• Secured 98.99 percentile in Computer Science and Engineering GATE – 2004 (Graduate Aptitute Test in Engineering)
• Secured state rank of 195 in EAMCET – 1998 (Engineering and Medical Common Entrance Test), Andhra Pradesh
• Awarded 1st Prize for Oral presentation at MEDBTCR’09 National Conference on Medical Biotechnology and Clinical Research (held at IISc, Bangalore)
• Travel fellowship from BioSapiens for attending ISMB/ECCB-2009

Co-curricular activities

• Worked in various student administrative committees at IISc
• Active member of Samskrit-sangha (Sanskrit awareness body at IISc)
• Learning and promoting martial arts (Kung-fu)
• An amateur mathematician – proposing conjectures and proving interesting theorems
• Practising classical music (Flute)

Research Publications

International Journal Articles

• Kalidas Yeturu and Chandra N (2008).“PocketDepth: a new depth based algorithm for identification of ligand binding sites in proteins”.

http://proline.physics.iisc.ernet.in/pocketdepth Journal of Structural Biology, 161, 31-42. Brief Description – PocketDepth employs a 3D grid to encompass protein and access sub-spaces. Each grid-cell is tagged with DepthFactor indicating number of grid-bars passing through it, drawn between pairs of surface-atoms. The tagged and proximal grid-cells are clustered based on density and DepthFactors. The resulting clusters are rank-ordered by size, polarity and other properties. The algorithm has been thoroughly validated on knowledge databases and is cited by 13 articles.

• Kalidas Yeturu and Chandra N (2008).“PocketMatch: a new algorithm to compare binding sites in protein structures”.

http://proline.physics.iisc.ernet.in/pocketmatch BMC Bioinformatics, 9:543 (17 pages). Brief Description – A frame-invariant shape signature for a binding site is derived by partitioning the set of all pair distances between 3D points into 90 sorted lists. These lists capture geometry and chemical nature of a site by assigning a residue to one of 5 chemical groups and representing it by 3 types of points. A pair of sites is compared by aligning corresponding distance lists and counting number of common elements. Novel validation strategies have been used for testing the algorithm and the method is cited by 10 articles.

• Karthik Raman, Kalidas Yeturu and Nagasuma Chandra (2008).“targetTB: A target identification pipeline for Mycobacterium tuberculosis through a reactome, interactome and genome-scale structural analysis”.

http://proline.physics.iisc.ernet.in/targetTB BMC Systems Biology,2:109 (44 pages) Brief Description – Systems biology methods including network analysis and Flux Balance Analysis and Essentiality studies (work by Dr. Raman) resulted in putative targets for M tb. My work involved estimation of targetability of pathogen proteins by matching their binding sites to those of the human proteins. The binding sites in host and pathogen are identified using PocketDepth and 245 million pairs of sites between the two organisms were compared using PocketMatch MPI version on IBM Bluegene. Number of citations is 20

• Kalidas Yeturu, Tapani Utriainen, Graham JL Kemp and Nagasuma Chandra (2010). “An automated framework for understanding structural variations in the binding grooves of MHC class II molecules”.

BMC Bioinformatics 11(Suppl 1):S55 doi:10.1186/1471-2105-11-S1-S55. (13 pages) Brief Description – A dataset of 103 MHC class II molecules is curated using 3D coordinates of 1DLH as template. Binding sites are extracted as peptide neighbourhoods and compared using PocketMatch13 algorithm that considers chemical and geometric distribution of atoms in the site. Site-cladograms were generated for whole-groove, P1, P4, P6, P7 and P9 sub-pockets and colour coded by disease types. Observed that P4 was found to be discriminatory

Reviews and Book chapters

• Karthik Raman, Kalidas Yeturu and Nagasuma Chandra (2008).“Model Driven Drug Discovery: Principles and Practices”, In: Biological Database Modeling.

Artech House INC, MA, USA (ISBN: 978-1596932586) Pg 163-188.

• Kalidas Yeturu and Nagasuma Chandra (2008). “Structural Bioinformatics: Transforming protein structures into biological insights”. Review,

Journal of the Indian Institute of Science, 88:2, 107-129.

• Karthik Raman, Kalidas Yeturu, Ashwini Bhat and Nagasuma Chandra (2009) Systems-level modelling of Mycobacterium tuberculosis for anti-tubercular drug target identification, Biobytes (Open access newsletter), 2009, 5:63-67
• Nagasuma Chandra, Praveen Anand and Kalidas Yeturu (2010). Structural Bioinformatics: Deriving biological insights from protein structures.

Interdisciplinary Sciences: Life Science. (In Press)

Manuscripts under preparation

• Kalidas Yeturu and Nagasuma Chandra, PocketAlign: An Algorithm for Aligning Binding sites in protein structures
• Kalidas Yeturu and Nagasuma Chandra. PocketDesign : A novel approach for designing binding sites
• Kalidas Yeturu, Praveen Anand Narvekar and Nagasuma Chandra, PocketAnnotate : A framework for functional annotation of binding sites.
• Kalidas Yeturu and Raghu Bhagavath and Nagasuma Chandra, LigandAlign : An algorithm for alignment of ligand structures
• Work on – classification of DHFR, MHC class-I and Nucleotide phosphate binding sites

International Conference presentations

• Kalidas Yeturu and Nagasuma Chandra (2006). CavityGrid: A new grid based algorithm for detecting binding cavities.

INCOB, Delhi, India.

• Kalidas Yeturu and Nagasuma Chandra (2008). Drug Targetability estimation through comparative ’pocket-omics’ of host and pathogen: A case study in Mycobacterium.

International Conference on Systems Biology (ICSB), Gothenburg, Sweden.

• Kalidas Yeturu, Karthik Raman and Nagasuma Chandra (2008). Targetability of Mycobacterium tuberculosis: Comparison of host and pathogen pocketomes.

Human Genome Meeting (HGM), Hyderabad, India.

• Kalidas Yeturu, Karthik Raman and Nagasuma Chandra (2009). Structural comparison of host and pathogen pocketomes – Drug targetability assessment in M. tuberculosis.

Joint Indo-Brazil Symposium on Macromolecular Crystallography, Indian Institute of Science, Bangalore.

• Kalidas Yeturu and Nagasuma Chandra (2009). Determining protein-ligand ’interaction space’ in the druggable portion of Mycobacterium tuberculosis proteome.

Special Interest Group (3DSIG), Stockholm, Sweden.

• Kalidas Yeturu, Graham J.L. Kemp, Nagasuma Chandra (2009). Deriving binding site signatures in MHC Class II molecules, using a novel algorithm.

ISMB/ECCB 2009, Stockholm, Sweden.

• PUBLICATION (4) was presented at APBC-2010, The Eighth Asia Pacific Bioinformatics Conference, Bangalore, India, 18 – 21 January 2010.

National conferences

• Kalidas Yeturu 2009 Structure based large scale prediction and comparison of binding pockets towards understanding drug specificity.

MEDBTCR’09 National Conference on Medical Biotechnology and Clinical Research, Bangalore, India.

• DST Sponsored SERC School on Machine Learning and Pattern Evolution methods in Chemo and Bioinformatics

26-29 June 2007 Jointly organized by National Chemical Laboratory (NCL), Pune Centre for Development of Advanced Computing (C-DAC), Pune.

• 11th ADNAT Convention - A Three Day Symposium on, Advances in Structural Biology and Structure Prediction. February 23, 24 & 25, 2007 Organized by CMB, Hyderabad, India.

Personal

• Born : 02-JUNE-1981
• Place of birth : Nellore, INDIA

Though register says June, I am a Saggitarian :-)