DEEPAK BANDYOPADHYAY
|
Current Address: |
Contact Information: |
|
|
Johnson
& Johnson Pharmaceutical R&D |
Cell: (919) 225-7711 dbandyop@prdus.jnj.com |
Work: (610) 458-5264 x6621 |
Summary
Currently postdoctoral research fellow at Johnson
& Johnson Pharmaceutical R&D, developing advanced cheminformatics and computational
chemistry algorithms for molecular modeling, conformational analysis, virtual
screening and 3D database search applied to drug discovery.
Recent Computer Science Ph.D. graduate from UNC Chapel
Hill, with broad experience and expertise in the application of computational
geometry and data mining techniques to structural bioinformatics. Key research contributions are a
geometric technique that makes nearest neighbor analysis of protein structures
more robust, and development of many new applications including robust packing
scores, graph mining to find family-specific structural motifs, function
inference for structural genomics, and analysis of conformational change.
Background in graphics, strong visualization
skills. Strong software
development abilities in C++ and Java, prototyping and numerical analysis in
MATLAB and scripting in Unix/Perl, with industry experience. Detailed knowledge of protein structure
and function from courses, independent study, and collaboration with domain
experts.
Education
University of North Carolina at Chapel Hill
Ph.D. in Computer Science, September 2005
M.S. in
Computer Science, May 2002
Indian Institute of Technology, Delhi, India
B.Tech.
in Computer Science and Engineering, Aug 1999, GPA=3.56/4.0
Skills
Subjects:
Computational Chemistry and Cheminformatics; Biogeometry (Computational
Geometry in Molecular Biology); Computational Biology and Bioinformatics
(especially Structural Bioinformatics); Computational Geometry; Data Mining;
Algorithm Design; Statistical Methods; Computer Graphics and Visualization;
Virtual and Augmented Reality
Languages: C, C++, Java,
MATLAB, scripting(Unix, Perl), HTML.
Software released (in C++): Almost-Delaunay
Simplices in CGAL, June 2004; gzstream library (with Lutz Kettner), Sep. 2001;
Dynamic Shader Lamps (with Ramesh Raskar), Aug. 2001
Technologies and APIs:
Computational Geometry (CGAL, LEDA); Databases and Data Mining
(ODBC/JDBC, FFSM subgraph mining, SVM); Graphics and Multimedia (OpenGL,
Renderman, Java3D, Swing, VRML, DirectMedia SDK); Vision and Image Processing
(OpenCV, Matrox Imaging Library);
Generic programming (STL, Boost); Parallel programming (MPI, OpenMP)
Bioinformatics and Cheminformatics Tools: Molecular modeling (Schrödinger, Sybyl, Accelrys,
DeepView, VMD, Kinemage, Pymol, Spock); Data Warehousing
(ABCD/3DX/DirectedDiversity); Standard bioinformatics sequence and structure
methods and web services (particularly: BLAST, PSI-BLAST, ClustalW, PROSITE,
InterPro, PFam, DSSP/DSSPcont, PROMOTIF, SCOP, CATH, VAST, DALI, PDB, PDBSum,
EC, GO, MolMovDB, SFLD, PINTS, SNAPP)
Development Tools and Platforms: Windows (Visual C++ .NET, Intel Compiler, JBuilder,
Eclipse), Linux and Unix variants (variety of compilers, editors, debuggers,
and other tools)
Other Software Packages: Geometric modeling (3D studio MAX, Rhino3D, Maya),
Image manipulation (Paint Shop Pro, Photoshop, GIMP), Video editing (Adobe
Premiere)
Experience
Postdoc, Johnson & Johnson Pharmaceutical Research &
Development, 10/2005-present
Developing computational chemistry algorithms
with the Molecular Design and Informatics group (now part of Research and Early
Development Information Management / Informatics). Development of 3D geometric algorithms
for conformational analysis, molecular alignment, pharmacophore development, 3D
database search and virtual screening, to be deployed within ABCD, an in-house
data warehousing software for cheminformatics and bioinformatics used by
Johnson & Johnson’s drug discovery scientists worldwide. (supervisor: Dr.
Dimitris K. Agrafiotis).
Research Assistant, University of North Carolina at Chapel Hill, 6/2001-9/2005
Ph.D.
Thesis: Developed a
geometric framework for robust nearest neighbor analysis in imprecise
points (Almost-Delaunay
simplices), and some of its applications in analysis of protein
structure: scoring functions for protein packing; secondary structure
determination; subgraph mining to find family-specific spatial motifs for
functional annotation of structural genomics targets; and analysis of
conformational change. (advisor: Dr. Jack Snoeyink).
Instructor, University
of North Carolina at Chapel Hill, 6/2003-7/2003
Designed and taught an undergraduate course on
introductory programming with Java
Summer Intern,
BIWI (Computer Vision Lab),
ETH Zurich, Switzerland. 6/2002-8/2002
Implemented the framework for a simultaneous
multi-projector and multi-camera structured light system which was used for
real-time capture of the human body.
Research Assistant, University of North Carolina at Chapel Hill, 8/2000-6/2001
Developed Dynamic Shader Lamps,
an interactive Augmented Reality system for painting on moving real objects
using projectors; demonstrated at ISAR’01 conference in New York.
Summer Intern,
HRL Labs, Malibu, CA. 6/2000-8/2000
Wrote applications for synchronized recording of video
and sensor data streams for use with a simulation testbed for visualizing
occlusions in an outdoor Augmented Reality system.
Research Assistant, University of North Carolina at Chapel Hill, 8/1999-5/2000
Explored occlusion resolving methods for Augmented
Reality in ultrasound-guided biopsy.
Summer Intern, Kent Ridge Digital Labs, Singapore. 6/1998-7/1998
Built GUI application
for visual feature extraction and feature-based query on a video database,
using Java, which was integrated into KRDL’s video indexing software.
Academic Projects
Graduate:
§
Kinemage
visualization of almost-Delaunay tetrahedra: Developed MATLAB code to generate
kinemage files for visualization of AD tetrahedra, packing, secondary
structures, spatial motifs and hinges. Course
project, Structure of Biological Macromolecules.
§
Structured light for 3D
object localization: Projector-camera system for real-time
tracking of a moving object, such as the user’s hand. Course project, Advanced
Image Synthesis.
§
Distance functions for
geometric simplification: Studied
effects of choosing a non-Euclidean distance function on a geometric
simplification method based on edge collapsing with a quadric error metric
(with M. Gopi). Course project, Geometric
and Solid Modeling.
§
Stained-glass mosaic
shader: Designed a Renderman
shader and designed ray-traced scenes with stained-glass mosaic effects and
translucency. Project, Advanced Image Synthesis.
§
Parallel N-body
simulations: Implemented
work-efficient parallel simulations using MPI and OpenMP on a 64-processor
machine. Course Project, Parallel and
Distributed Computing.
§
Robust Internet audio
codec: Implemented system for low
bitrate transmission of real-time audio in the presence of simulated load and
noise. Course Project, Multimedia
Networking.
Undergraduate: Distributed
Virtual Environment for multi-user control of an articulated figure (thesis
project); Face recognition; Simulation of network congestion control schemes;
Distributed matrix multiplication/exponentiation; Slot machine simulation on
FPGAs.
Publications
Refereed journal
papers
Deepak
Bandyopadhyay and Dimitris K Agrafiotis. A New Self-Organizing Algorithm for Molecular
Alignment and Pharmacophore Development. Journal of Chemical Information and Modeling,
submitted.
Dimitris K Agrafiotis, Deepak
Bandyopadhyay, Giorgio Carta, Andrew J S Knox and David G Lloyd. On the Effects of
Permuted Input on Conformational Sampling: An Evaluation of Stochastic
Proximity Embedding (SPE).
Chemical Biology & Drug Design, to appear.
Dimitris K Agrafiotis, Deepak
Bandyopadhyay, Joerg K Wegner and Herman van Vlijmen. Recent Advances in
Chemoinformatics. Journal of Chemical Information and Modeling, 2007, to
appear.
Dimitris K Agrafiotis, Deepak
Bandyopadhyay and Michael Farnum. Radial
Clustergrams: Visualizing the Aggregate Properties of Hierarchical
Clusters. Journal of Chemical
Information and Modeling, 2007, vol. 47, pp. 69-75.
Deepak Bandyopadhyay and Jack Snoeyink. Almost-Delaunay Simplices : Robust
Neighbor Relations for Imprecise 3D Points Using CGAL. Journal of Computational Geometry
Theory and Applications (CGTA), 2007, to appear.
Deepak Bandyopadhyay, Jun Huan, Jinze Liu, Jan Prins, Jack Snoeyink, Wei Wang, and
Alexander Tropsha. Structure-Based Function Inference Using Family-Specific
Fingerprints. Protein Science, vol. 15, June 2006, pp. 1537-1543.
Jun Huan, Deepak
Bandyopadhyay, Wei Wang, Jack Snoeyink, Jan Prins and Alexander Tropsha. Comparing
Graph Representations of Protein Structure for Mining Protein Family-Specific
Residue Packing Motifs. Journal of Computational Biology (JCB), Vol. 12,
No. 6, July 2005, pp. 657-671.
Refereed conference
and workshop papers
Jun Huan, Deepak
Bandyopadhyay, Jack Snoeyink, Jan Prins, Alexander Tropsha and Wei
Wang. Distance-based
Identification of Spatial Motifs in Proteins Using Constrained Frequent
Subgraph Mining. Proceedings of the IEEE Life Sciences Society
Computational Systems Bioinformatics (CSB) conference, Aug. 14-18, 2006.
Deepak Bandyopadhyay, Jun Huan, Jinze Liu, Jan Prins, Jack Snoeyink, Wei
Wang, and Alexander Tropsha. Structure-Based Function Inference Using
Family-Specific Subgraph Fingerprints Mined from Protein Families.
Proceedings of the first ISMB Special Interest Group meeting on Automated
Function Prediction (AFPSIG), 2005.
Deepak Bandyopadhyay
and Jack Snoeyink. Almost-Delaunay Simplices : Robust Neighbor Relations
for Imprecise 3D Points Using CGAL. 2nd CGAL User Workshop, June 2004.
Jun Huan, Wei Wang, Deepak Bandyopadhyay, Jack
Snoeyink, Jan Prins and Alexander Tropsha. Mining Protein Family-specific
Residue Packing Patterns from Protein Structure Graphs. International
Conference on Research in Computational Molecular Biology (RECOMB), March 2004,
pp. 308-315.
Deepak Bandyopadhyay
and Jack Snoeyink. Almost-Delaunay Simplices: Nearest Neighbor Relations
for Imprecise Points. ACM-SIAM Symposium on Discrete Algorithms (SODA),
Jan. 2004, pp. 403-412.
Graphics:
Deepak
Bandyopadhyay, Ramesh Raskar and
Henry Fuchs. Dynamic Shader Lamps: Painting on Real Objects.
International Symposium on Augmented Reality (ISAR), Oct. 2001.
Ramesh Raskar, Greg Welch,
Kok-Lim Low and Deepak Bandyopadhyay. Shader Lamps: Animating Real
Objects with Image-Based Illumination. Eurographics wkshop(EGWR), 2001.
Non-refereed
publications (technical reports, posters and oral presentations)
Deepak Bandyopadhyay. Advances in Conformational Sampling
Methods: Comparative Studies and Application to Pharmacophore Development. North
American MOE User Group Meeting, Montreal, Canada, June 2007.
Deepak Bandyopadhyay. Target Identification and
Target-Specific Drug Discovery Using Protein Family-Specific Fingerprints.
Mid-Atl. Reg. ACS Mtg.(MARM’07), Collegeville, PA.
Deepak Bandyopadhyay.
New Self-Organizing Algorithm for Molecular Alignment and
Pharmacophore Development. Talk at 233rd ACS meeting, Chicago,
Mar 2007 and eCheminfo 2006, Oct. 2006, Bryn Mawr, PA.
Deepak Bandyopadhyay, Jun Huan, Jinze Liu, Jan Prins, Jack Snoeyink and Wei
Wang. Function Inference Using
Family-Specific Subgraph Fingerprints Mined From Protein Families. Poster presented at Intelligent
Systems for Molecular Biology (ISMB) 2005, June 2006, Detroit, MI.
Deepak Bandyopadhyay, Jack Snoeyink, Alexander Tropsha and Charlie Carter. Analysis of Protein Conformational Change
Using Almost-Delaunay Tetrahedra. Poster presented at Pacific Symposium on
Biocomputing (PSB), Jan. 2005, Big Island of Hawaii.
Deepak Bandyopadhyay, Jun Huan, Jinze Liu, Jan Prins, Jack Snoeyink and Wei
Wang. Using Fast Subgraph Isomorphism for Protein Function Annotation Using
SCOP and Gene Ontology. UNC-CS Technical Report TR04-031, 2004.
Deepak Bandyopadhyay, Jun Huan, Wei Wang and Jack Snoeyink. A
Fast Subgraph Search Scheme Using a Local Index for Functional Annotation of
Protein Structures. Poster presented at Intelligent Systems for Molecular
Biology (ISMB-ECCB’04), Aug. 2004,
Glasgow, UK.
Deepak Bandyopadhyay. Graph Representations and Algorithms for Protein
Family Classification and Functional Annotation. Laptop Presentation at the Special Interest Group
meeting in Structural Bioinformatics (3DSIG), ISMB-ECCB’04, July 2004, Glasgow,
UK
Deepak Bandyopadhyay, Alexander Tropsha and Jack Snoeyink. Analyzing
Protein Structure using Almost-Delaunay Tetrahedra. UNC-CS Technical Report
TR03-043, 2003. Poster presented
at RECOMB 2004, March 2004, San Diego, CA.
Deepak Bandyopadhyay, Ramesh Raskar, Andrei State
and Henry Fuchs. Dynamic
Spatially Augmented 3D Painting. UNC-CS Technical Report TR01-006, 2001.
Invited Talks (other than conference presentations)
§
National Center for
Biotechnology Information, Bethesda, MD, Feb. 2005
A geometric framework for
robust neighbor analysis of protein structure and function
§
European Bioinformatics
Institute, Cambridge, UK, July 2004
Methods and applications for
nearest neighbor analysis of protein structures
§
Duke Algorithms Seminar,
Duke University, Durham, NC
o
Almost-Delaunay
tetrahedra for analyzing protein structure, March 2003
o
Mining family-specific residue
packing patterns from protein structure graphs
(with Luke Huan), Nov. 2004
§
Biogeometry PI meetings
o
Computing
Almost-Delaunay triangles and edges, Stanford University, Oct. 2001
o
Structure-based function
inference using family-specific subgraph fingerprints mined from protein
families, Big Island of Hawaii, Jan. 2005
Professional Activities and
Affiliations
Reviewer for ISMAR (augmented
reality), CIKM and KDD (databases and data mining),
Computational Geometry: Theory and Applications, and Bioinformatics
Current or past member of ISCB
(International Society for Computational Biology), ACS (American Chemical
Society)
Awards
§
Secured the 49th
rank among over 100,000 applicants in Joint Entrance Exam for the IITs.
§
Qualified for the
Singapore Airlines (SIA-NOL) scholarship, 1995.
§
Received merit prize
(among top 25 students in class of
375) for 2 semesters at IIT Delhi.
§
Secured a rank amongst
the top 1% (all India) in National Physics Olympiad 1994-95.
§
National Talent Search
Examination (NTSE) scholarship, 1993.
Travel, creative writing,
trivia and word games, crossword creation, digital photography
Language skills: English, Hindi (native speaker); French, Bengali
(conversational)
References
§
Dr. Dimitris K. Agrafiotis,
Vice President, Research and Early Development Information Management
(Informatics), Johnson & Johnson Pharmaceutical R&D, Exton, PA.
Phone: (610) 458 6045 Email: dagrafio @ prdus.jnj.com
§
Dr. Jack Snoeyink,
Professor, Dept. of Computer Science, UNC Chapel Hill.
Phone: (919) 962 1969 Email: snoeyink @ cs.unc.edu
§
Dr. Alexander Tropsha,
Professor and Director, Laboratory of Molecular Modeling, Dept. of Medicinal
Chemistry and Natural Products, School of Pharmacy, UNC Chapel Hill.
Phone: (919) 966
0204 Email: alex_tropsha @ unc.edu
§
Dr. Wei Wang, Assistant
Professor, Dept. of Computer Science, UNC Chapel Hill.
Phone: (919) 962 1744 Email: weiwang @ cs.unc.edu