Bioinformatics Methodologies

People

Purdue University, Department of Statistics, Dr. Jayanta K. Ghosh
Wroclaw University of Technology, Poland, Institute of Mathematics, Dr. Malgorzata Bogdan
Carnegie Mellon University, Department of Statistics, Dr. Surya Tokdar
Purdue University, Department of Statistics, Mr. Ryan Martin

Description

In some of our recent work we have tried to develop new methodology for multiple tests or variable selection in high dimensional problems. Our goal has been to clarify and develop insight into problems of finding significant markers in QTL studies or micro-array data. Through simulation and theoretical studies, we have developed new methods, e.g., variable selection by boosting and mBIC [6, 5], extension of our mBIC to dense markers and multiple interval mapping [4]. We have also introduced new Nonparametric Bayes and Nonparametric Empirical Bayes methods suitable for micro-array data and compared them with an oracle in [3]. The oracle has been developed at the same time as similar new work by Stoney in JRSS Ser B (2007).

These and other new relevant work on estimating mixing distributions in mixture models are listed below. We hope to extend these methods for identifying gene pathways from micro-array data.

Several problems in Bioinformatics are discussed in [9].

References:

(With Junyong Park) Persistence of Plug-in Rule in classification of High Dimensional Multivariate Binary Data, JSPI vol 137, pp 3687-3705.
(With M. Bogdan, A. Ochman, S. Tokdar) (2007) On the Empirical Bayes Approach to the Problem of Multiple Testing, Quality and Reliability Engineering International (a Wiley Journal). vol 23, pp 727-739.
(With M. Bogdan, S. Tokdar) (2008) A Comparison of the Benjamini-Hochberg Procedure with some Bayesian rules for Multiple Testing, in IMS Lecture Note series, N. Balakrishnan et al eds. IMS Collections, vol 1, pp 211-230.
(With M. Bogdan, F. Frommlet, P. Biecek, R. Cheng, R. Doerge) (2008) Extending with Modified Bayesian Information Criteria (mBIC) to dense Markers and Multiple Interval Mapping, 1 Biometrics 2008 (to be printed, available online).
(With M. Bogdan, M. Zak-Szatkowska) (2008) Selecting Explanatory Variables with the Modified Version of Bayeian Information Criterion, Quality and Reliability Engineering International (2008, to be printed).
(With J. Park, J. Wilbur, C. Nakatsu, C. Ackerman) (2007) Selection of Ginavy Variables and Classification by Boosting, Communications in Statistics: Simulation and Computation, vol 36, pp 855-869.
(With R. Martin) (2008) Stochastic Approximation and Newton's Estimate of a Mixing Distribution, Statistical Science (to appear).
(With S. Tokdar, R. Martin) Consistency of a Recursive Estimate of Mixing Distributions, Annals of Statistics (to appear).
(With M. Bogdan and T. Samanta) (2008) Applications and Indianness of Indian data (invited paper to appear in Sankhya Series B).