Nitin Gupta @ UCSD
 
Search this site

Navigation

Projects

This page is under construction.


Bioinformatics - Mass Spectrometry (my thesis work)

Identification of neuropeptides and proteolytic pathways using mass spectrometry
(Advisors: Pavel Pevzner and Vivian Hook)

We are trying to identify novel neuropeptides and processing mechanisms in humans and cow using tandem mass spectrometry and computational analysis.

Rigorous computation of false discovery rates of peptide and protein identifications.
(Advisor: Pavel Pevzner)

Computation of error rates by various database search algorithms has been a messy field so far. Our work provides a rigorous framework for these computations and removes the need for decoy databases.
  • N. Gupta, N. Bandeira and P.A. Pevzner. Target-decoy approach in proteomics: when things may go wrong. Submitted.
  • N. Gupta and P.A. Pevzner (2009). False discovery rates of protein identifications: a strike against the two-peptide rule. Journal of Proteome Research (accepted).
  • S. Kim, N. Gupta, N. Bandeira and P.A. Pevzner (2009). Spectral Dictionaries: Integrating De Novo Peptide Sequencing with Database Search of Tandem Mass Spectra. Molecular and Cellular Proteomics. 8(1):53-69.
  • S. Kim, N. Gupta and P.A. Pevzner (2008). The Partition Function of Tandem Mass Spectra: a New Approach to Peptide Identifications. Journal of Proteome Research. 7(8): 3354 - 3363.
Proteogenomics and Comparative Proteogenomics
(Advisors: Pavel Pevzner, Richard D. Smith, Vineet Bafna)

We have developed algorithms and software tools for using mass spectrometry data for improving gene annotations, finding proteolytic sites, post-translational modifications and operons.
  • L. Wich and N. Gupta. MS-Operon: Using tandem mass spectrometry for operon prediction and validation. Submitted.
  • N. Gupta, J. Benhamida, V. Bhargava, D. Goodman, E. Kain, I. Kerman, N. Nguyen, N. Ollikainen, J. Rodriguez, J. Wang, M.S. Lipton, M. Romine, V. Bafna, R.D. Smith and P.A. Pevzner (2008).  Comparative Proteogenomics: Combining Mass Spectrometry and Comparative Genomics to Analyze Multiple Genomes.
    Genome Research.     18:1133-1142
  • J. Rodriguez, N. Gupta, R.D. Smith and P.A. Pevzner (2008). Does trypsin cut before Proline? Journal of Proteome Research. 7(1):300-5.
  • N. Gupta, S. Tanner, N. Jaitly, J.N. Adkins, M. Lipton, R. Edwards, M. Romine, A. Osterman, V. Bafna, R.D. Smith and P.A. Pevzner (2007).  Whole proteome analysis of post-translational modifications: applications of mass-spectrometry for proteogenomic annotation. Genome Research. 17(9):1362-77.

Bioinformatics - Others

To be added




Neuroscience/Cognitive Science


Computational modeling of Remote Associates Test
(Advisors: David Huber and Sara Mednick)

We are working on finding the normative and descriptive models for the Remote Associates Test, which has been used as a test of creativity for over 40 years. This work started as a class project in Dave's class on Mathematical Models in Psychology.




Others (older projects during my undergrad)

Monte Carlo study of coupled folding-binding of proteins on a lattice model
(Advisor: Prof. Anders Irback)
    The idea was to see if coupling of folding with binding changes the kinetic and thermodynamic properties of binding. We did observe some interesting deviations from regular docking, suggesting that structural flexibility might actually be an important trait for some proteins which are unstructured in isolation.

Chinese Checkers
    Designed and implemented an algorithm for this multiplayer board-game that can be played by 2, 3, 4 or 6 players, using a modified alpha-beta Minimax Game Tree. Depth of the tree was kept short to make the program faster and pruning at various levels and in varied amounts was done to optimize the performance. This work was done under the guidance of Prof. Manindra Agrawal, and my team received the BEST PROJECT award for the winning the knockout competition between 20 such implementations of the game.

Two-Criterion optimization in state assignment for synchronous Finite State Machines.     A challenging problem in micro-electronics is finding the best state assignment for implementing a synchronous sequential circuit which are also represented as Finite State Machines. Two conflicting objectives are considered while finding optimal assignments: reducing the number of encoding bits, and having a low cost of transition across states. We use  multi-objective evolutionary algorithms to find the pareto optimal set of solutions. This work was done as a course project for the course "Multi-Objective Optimizations" taught by Prof. Kalyanmoy Deb and Dr. J. Dutta.

Discovery of global associations in the amino-acid sequences of proteins.
    We applied the techniques of quantitative association rule mining to decipher the statistically significant co-occurrence of amino acids across the protein sequences. This project was done as a requirement for the course "Data Mining", taught by Dr. Pabitra Mitra.

Protein contact map prediction and evaluation of structural similarity using physical and graph theoretic properties measurable through contact maps. We developed a genetic algorithms based approach for finding an approximate contact map to protein structure, using a feasibility measure based on the satisfaction of physical constraints. This work, done with my partner Nitin Mangal in the guidance of Prof. Somenath Biswas, was nominated as a candidate for the best B. Tech. Project in the department.

Distinguishing proteins from random sequences.  We addressed the question whether protein sequences have significant statistical deviation from randomness, and developed a neural network approach to classify a given amino acid chain as protein or random sequence based on a set of 27 carefully chosen sequence-related parameters. Done under the guidance of Prof. Harish Karnick.