David McCoy

David McCoy

PhD Candidate in Environmental Health Sciences

University of California, Berkeley

Biography

I’m a PhD candidate in the Division of Environmental Health Sciences at the University of California, Berkeley under the supervision of Alan Hubbard and Martyn Smith. My work revolves around developing assumption-lean methods for the statistical analysis of mixed exposure data. That is, in most realistic exposure settings humans are exposed to a mixture of chemicals which modify biological pathways and lead to disease states. I develop statistical methods to understand what parts of a mixed exposure cause disease and through which pathways. I collaborate with epidemiologists, biologists, and medical doctors to develop sound statistical approaches to answer real world problems.

Interests
  • Data adaptive target parameters
  • Mixed Exposures
  • Mediation
  • Nonparametric and semiparametric inference
  • Computational biology
  • Statistical software development
Education
  • PhD in Environmental Health Sciences, 2019-present

    University of California, Berkeley

  • MSc in Epidemiology, 2015-2018

    London School of Hygiene and Tropical Medicine

  • BA/BSc in Philosophy, Cognitive Science and Psychology, 2009-2014

    University of Delaware

Recent Publications

Convolutional Neural Network–Based Automated Segmentation of the Spinal Cord and Contusion Injury: Deep Learning Biomarker Correlates of Motor Impairment in Acute Spinal Cord Injury
Texture features from computed tomography correlate with markers of severity in acute alcohol-associated hepatitis
Predicting the binding of small molecules to nuclear receptors using machine learning
Endoluminal biopsy for molecular profiling of human brain vascular malformations

Teaching

University of California, Berkeley

Selected Experiences

 
 
 
 
 
Epidemiologist
University of California, San Francisco
Oct 2014 – Oct 2018 San Francisco, CA
Worked with medical doctors to develop research plans for study development. Generally worked as an applied statistician on a variety of projects ranging from imaging to genetics. Perfomed analyses and co-wrote publications and grants.
 
 
 
 
 
Data Scientist
University of California, San Francisco
Oct 2014 – Mar 2020 San Francisco, CA
Developed novel machine learning methods for the analysis of high-dimensional data. Conceived and wrote several neural network archetectures for the analysis of spinal cord injury, brain hemorrhage, hepatitis and other diseases.
 
 
 
 
 
Graduate Student Researcher
Oct 2019 – Present Berkeley, CA, United States
Develop assumption-lean statistical inference methods for mixture effect discovery. Create new semi-parametric definitions for interaction, effect modification and mediation using stochastic intervention.