Subha Guha is an expert in Bayesian biostatistical modeling for cancer genomics and computing for high-dimensional datasets. As PI or co-Investigator of research grants from NIH and NSF, he has developed novel Bayesian models for multi-domain, high-throughput biomedical studies. He has extensive experience with statistical computing to efficiently implement these methodologies for Big Data.

The primary focus of his research has been the development of broadly applicable, nonparametric statistical methodologies that are flexible because they avoid making unrealistic assumptions about the data features and permit nonlinear dependencies; scalable, because the procedures are capable of accommodating the ever-expanding massive, multiple-domain datasets, even on a modest computing budget; and most importantly, scientifically interpretable, because they are based on models that incorporate domain knowledge and provide meaningful answers to key scientific questions that motivate the research.

• Bayesian modeling
• Biostatistics
• Causal inference
• Clustering and classification
• Cognitive Neuroscience
• Computational methods for Big Data
• Connectomics
• Development and testing of novel medical devices
• Generalized linear models
• Health disparities and vulnerable populations
• Hidden Markov models
• High-dimensional inference
• MCMC simulation
• Microbiome data analysis
• Nonparametric Bayesian methods
• Observational studies
• Process monitoring
• Quality control
• Spatial statistics
• Statistical computing
• Statistical methods
• Survival Analysis
• cancer genomics
• electronic health records