Statistical Analyses of multiple omics dataEvaluate informatic tools for multiple omics integration

I expatriated to Vellore, India for 6 months in the early part of 2015 to organize and supervise a clinical field study on the microbiome of obese and non-obese child and mother pair at the Christian Medical College. This pilot study is being conducted because India faces two seemingly contradictory epidemics a) malnutrition and b) a growing prevalence of obesity with the introduction of Western diets. I aided in training field workers who measured anthropometrics and collected stool samples for 16S next generation sequencing. I also have two other ongoing projects. The first is a studying comparing the microbiome of HIV infected patients on antiretroviral therapy (ART) compare to negative controls. This study is similar to my previous study with the exception of having a much larger sample size. The second is a longitudinal study of the microbiome of children with Cryptosporidium and rotavirus before and after taking the probiotic Lactobacillus.All samples being sequenced target the V4 hypervariable region for the Illumina Miseq. Bioinformatics and statistics will be perform on QIIME and R Statistics.All three studies are currently under different stages of completion.

I compared the differences between the microbiome of a) HIV positive subjects who were on antiretroviral therapy (ART) to HIV negative controls around the greater Boston area in this cross-sectional study and b) stunted and non-stunted children from 3 months to 24 months of age from the Vellore district in Tamil Nadu, India for this longitudinal study. The V3-V5 hypervariable regions of the 16S were sequenced using the Roche 454 platform in the HIV study, and the V4 region was sequenced using the Illumina MiSeq platform in the stunting study. Bioinformatics and statistics were done on QIIME and R Statistics respectively. Our findings for the HIV study was published on January of 2015 in the Journal of Infectious Diseases. We are on the final revision for publication in PLOS ONE for our study in India.

This research on microbial evolution investigated a genetic network using E. coli as a paradigm. We determined how five mutations interacted (epistasis) and how these interactions affect fitness and viability. Several methods such as transduction and transformation were used to introduce the individual mutation and different mutational combinations into an ancestral background. The fitness of the new strains were then compared to the ancestral strain to assess if the combination of mutations gave the expected fitness benefit from which all mutations were individually previously measured. Our findings were publish in Science in 2011.During this time, I also mentored undergraduate and graduate students in experimental design and proper lab techniques using other microbial systems.