I am interested in developing novel methods and improving existing methods for exploring and analyzing interesting patterns in complex data from different application domains. My current projects are in the fields of statistical genetics, bioinformatics and computational biology, feature selection and classification, and social network analysis.
Selected projects on this topic: Nonparametric feature selection. Evaluation of reproducibility of studies with large number of objects.
In projects on this topic, we tackle the challenging problems of identifying pre-disposing genes for complex human disorders.
Project with website: the application of BHTA to the Inflammatory Bowel Disease (2003-2004)
Bioinformatics: gene expression profiling
Selected projects on this topic: selecting genes with interactions for tumor classification; tree-based classification for multi-class gene expression data; genetics of gene expression.
Selected projects on this topic: phylogenic modelling, graph models for molecular pathways.
This project is trigged by a special survey asking people questions like “how many X do you know”. Selected projects on this topic: measuring social structure using overdispersion; efficient estimation of personal network size; estimating latent non-random mixing of social network; latent space models for aggregated relational data.
Project with website: “How many people do you know in prison?” (2003-2004)
The purpose of this project is to develop and evaluate web-based teaching tools on conducting data project for students enrolled in an intro-level statistics class.