Research

Modern computing has greatly accelerated the collection of large amounts of data. Searching for important statistical patterns within this data is harder than simply finding a needle in a haystack. It is more akin to looking for a specific needle in a box of needles in a needle box factory.

My work focuses on developing statistical methodology to analyze data from the human microbiome and determine which microbial taxa are linked to common human health issues such as inflammatory bowel disease and obesity. I am interested in establishing methods that guard against common threats to non-reproducibility. Specifically, my work broadly focuses on methods that control the number of false positive taxa and methods that are explicitly robust to outliers within the data.

False Discovery Rate Control:

  1.  Arun Srinivasan (2017). Master’s Thesis: Calibrating Black Box Classification Models through the Thresholding Method. 

Robust Statistics:

  1.  Arun Srinivasan, Danning Li, Lingzhou Xue, and Xiang Zhan (Under Review 2020+) Robust Shape Matrix Estimation for High-Dimensional Compositional Data with Application to Microbial Inter-Taxa Analysis.

Hypothesis Testing:

  1.  Kalins Banerjee, Ni Zhao, Arun Srinivasan, Lingzhou Xue, and others (2019). An Adaptive Multivariate Two-Sample Test With Application to Microbiome Differential Abundance Analysis. frontiers in Genetics.

Additional Interdisciplinary Work:

  1.  Hannah Chazin, Soudeep Deb, Joshua Falk, and Arun Srinivasan (2018). New Statistical Approaches to Intra‐individual Isotopic Analysis and Modelling of Birth Seasonality in Studies of Herd Animals. Archaeometry.