Statistical and computational methods for high-throughput genomics. Computational epigenomics. Metagenomics. Second-generation sequencing data quality assessment and analysis.
I also work on general methods for machine learning and data science: kernel methods for predictive model building; mixed-integer programming for combinatorial estimation problems; interactive visualization; theoretical results on semidefinite programming for model selection; classical database query optimization techniques for querying probabilistic data.
More detail on my research projects may be found here, but some highlights include:
- Interactive genomic data visualization
- Anti-profiles: classifying cancer using heterogeneity
- Computational epigenomics
- Second-generation sequencing data quality assessment and analysis