current fellows (2022-2023)
Stephanie Armbruster
Academic advisor: Junwei Lu
Stephanie is a first year PhD student in the Biostatistics PhD program. Her primary research interest is clinical studies on genomics related questions, in particular how to tackle underrepresentation of marginalized groups.
During the 2022 fall term Stephanie is enrolled in the following courses: BST230: Probability I, BST 232: Methods, BST 254: Adaptive Clinical Trials, EPI 201: Introduction to Epidemiology I, BST 227: Introduction to Statistical Genetics, EPI 507: Genetic Epidemiology and HPM 548: Responsible Conduct of Research. And in the Spring 2023 term Stephanie plans to enroll in the following courses: BST 249: Bayesian Methodology in Biostatistics, BST 231: Statistical Inference I, BST 245: Analysis of Multivariate and Longitudinal Data, GENETIC 228: Genetics in Medicine – from Bench to Bedside, STAT 171: Introduction to Stochastic Processes, EPI 511: Advanced Population & Medical Genetics, and BST 234: Introduction to Data Structures & Algorithms.
Stephanie is planning to complete her first dry lab rotation in the spring 2023 term and her second dry lab rotation the following summer. The projects and supervisor are not decided at this time.
Dylan Clark-Boucher
Academic Advisor: Briana Stephenson
Dylan is a first year PhD student in the Biostatistics PhD program. His research interests include cancer genomics, rare genetic disease, and statistical methods for high-dimensional or multivariate data analysis.
During the 2022 fall term Dylan is enrolled in the following courses: BST 232: Methods, BST 254: Adaptive Clinical Trials, BST 262: Computing for big data, BST 280: Introductory genomics and bioinformatics for health research, EPI 217: Epidemiology of adult psychiatric disorders, EPI 249: Molecular biology for epidemiologists, and EPI 507: Genetic Epidemiology. And in the Spring 2023 term Dylan plans to enroll in the following courses: BST 249: Bayesian Methodology in Biostatistics, BST 231: Statistical Inference I, BST 234: Introduction to Data Structures & Algorithms, and HPM 548: Responsible Conduct of Research.
Dylan is planning to complete his first dry lab rotation in the spring 2023 term with Dr. Jeffrey Miller on a project involving XDP—a rare genetic disease—with the aim of developing statistical methods to properly evaluate disease progression by identifying relevant diagnostic variables. The project may have extensions to other, similarly rare genetic diseases whose progression has not been well-studied. Plans for his second dry lab, and wet lab rotations are still in the planning stages.
Christian Covington
Academic Advisor: Jeffrey Miller
Christian is a first year PhD student in the Biostatistics PhD program. His research interests are uncertainty beyond that from finite sampling (e.g. specification search, data processing, etc.). During the summer Christian started working on two research projects with Tyler VanderWeele. The first is about “multiverse (or specification curve) analysis”, a means of comparing statistical results across a variety of model specifications. This approach has become reasonably popular in social and medical sciences, but there is little theoretical and/or statistical work describing exactly what the method affords you, how you can analyze it, etc. and he is working on formalizing some of those ideas. The second is about joint influence estimation. Traditional diagnostics for coefficient estimation (e.g. DFBETAS) are typically limited to estimating the influence of single data points. However, it is known that these methods can do a poor job estimating joint influence (i.e. what would happen to coefficient estimates if multiple data points were removed). Recent theoretical work gives a method that estimates joint influence for OLS problems, but it is computationally infeasible if you have more than a few covariates. Christian is working on an approximate version of the method which generalizes beyond OLS and is computationally efficient.
During the 2022 fall term Christian is enrolled in the following courses: BST 232: Methods, EPI 201: Intro Epidemiology, HPM 548: Responsible Conduct of Research, STAT 221: Computational Tools for Statistical Learning, and STAT 286: Causal Inference. And in the Spring 2023 term Christian plans to enroll in the following courses: BST 249: Bayesian Methodology in Biostatistics, BST 231: Statistical Inference I, STAT 212: Probability II, STAT 220: Bayesian Data Analysis, and STAT 293: Design of Experimental and Non- experimental Studies.
Christian does not have definite plans for his rotations at this time but will likely complete dry lab rotations with Drs. Jeffrey Miller and Junwei Lu.
Rebecca Danning
Academic advisor: Xihong Lin
Rebecca is a second year PhD student in the Biostatistics PhD program. Her primary research interests are genetic subtypes of functional diseases.
During the 2022 fall term Rebecca is enrolled in the following courses: BST245: Analysis of Multivariate and Longitudinal Data, and STAT 286: Causal Inference with Applications. And in the Spring 2023 term Rebecca plans to enroll in the following course(s): EH 516: Environmental Genetics, EPI511: Advanced Medical and Population Genetics, BST 249: Bayesian Methodology in Biostatistics, BST 241: Statistical Inference II, and STAT 249: Design of Experimental and Non-experimental Studies.
Rebecca completed both of her dry lab rotations. The first dry lab rotation was with Professor Xihong Lin to analyze subtypes of Irritable Bowel Syndrome based on the UK Biobank digestive questionnaire database. She performed latent class analysis on the dichotomized digestive questionnaire variables and then performed common and rare variant analysis using the SAIGE and STAAR pipelines to look for genetic associations among the derived latent class phenotypes. In her second dry lab rotation Rebecca worked with Professor John Quackenbush to explore methods for data clustering. Ran simulations using network-based community detection algorithms as well as data science-based methods such as UMAP and tSNE. She is currently working on her wet lab rotation shadowing a postdoc in the lab of Professor Wendy Garrett. There she is learning about the bench science underlying genetic data, including passaging cell cultures, DNA extraction, and centrifuging samples with a focus on colon cancer and inflammatory bowel disease.
Raphael Kim
Academic advisor: Junwei Lu
Rapahel is a second year PhD student in the Biostatistics PhD program. His primary research interests are high dimensional statistics, reinforcement learning (RL), optimal treatment decisions and inference in these settings.
During the 2022 fall term Raphael is enrolled in the following courses: BST 227: Introduction to Statistical Genetics, EPI 217: Epidemiology of adult psychiatric disorders, MIT 9.S915: Special Subject in Brain and Cognitive Sciences. And in the Spring 2023 term Raphael plans to enroll in the following course(s): BST241: Statistical Inference II, BST 238: Advanced Topics in Clinical Trials, and SBS 202: Child Psychiatric Epidemiology.
Raphael has completed both of his dry lab rotations. The first, with Susan Murphy worked on assessing personalization in Reinforcement Learning (RL) settings using bootstrap. Post study inference following adaptive sampling clinical trials (finite time horizon T) does not have all the statistical tools built out for it. They investigated how to perform inference in this setting using bootstrap – in particular for answering questions related to personalization like how well did we learn on average or how many users did we really find running this algorithm? In his second rotation he worked with Dr. Rachel Nethery on causal inference and optimal treatment regimes in bipartite settings. Bipartite causal inference is a setting where you intervene on one unit type but observe outcomes on another unit type. This has applications in genomics, environmental health, and policy analysis. Raphael worked on simulations and showing statistical validity to a procedure based on an outcome aware regression model. Plans for his wet lab rotation are not finalized, but he tentatively plans to work with Dr. David Christiani.
Parker Knight
Academic advisors, Rui Duan
Parker is a second year PhD student in the Biostatistics PhD program. His primary research interests are statistical genetics and high dimensional statistics.
During the 2022 fall term Parker is enrolled in the following courses: EPI 249: Molecular Biology for Epidemiologists, BST 235: Advanced Regression and Statistical Learning. And in the Spring 2023 term Parker plans to enroll in the following course: BST241: Statistical Inference II.
Parker completed his first dry lab rotation with Dr. Rui Duan on multi-cohort polygenic risk score models using summary statistics. He was responsible for designing their model and developing theoretical guarantees for our estimator. Parker is currently working with Dr. Rajarshi Mukherjee on his second dry lab rotation that is looking at multi-phenotype association testing. He is responsible for a theoretical analysis of a submatrix detection algorithm for detecting multi-SNP multi-phenotype associations. He plans to complete his Dry Lab rotation during the spring 2023 term.
Corriene Sept
Academic advisor, Martin Aryee
Corriene is a third year PhD student in the Biostatistics PhD program. Her primary research interest is high-resolution mapping of transcription factors at DNA loop anchors.
During the 2022 fall term Corriene is enrolled in the following courses: EPI 249: Molecular Biology for Epidemiologists, and BST312A: Consultation.
Corriene has completed both of her Dry Lab rotations. Her first dry Lab rotation with Dr. Peter Kraft focused on building a polygenic risk score for VTE by building several different polygenic risk scores for VTE using summary statistics from the most recent GWAS. Corriene then evaluated the performance of the different models by applying them to individual-level data. The methods she used to make the PRSs are LDPred2-auto and SCT. In her second dry Lab rotation Corriene worked with Dr. Martin Aryee on applying existing loop- calling methods to MNase data (a new, higher resolution assay) to evaluate their performance. The goal is to develop a new method that is more suited to this new data type once they’ve evaluated the holes in existing methods. She plans to write and submit papers for both rotations. Corriene received permission to waive her wet lab rotation as she had extensive wet lab experience in college, and instead extended her project with Dr. Aryee.
Corriene is planning to submit her first dissertation paper with advisor Martin Aryee, “High resolution mapping of transcription factors at DNA loop anchors”, before spring semester 2023. Collaborators include Christian Cerda-Smith, Esther Tak, Haley Hutchinson, Kayla Oliveira, Viraat Goel, Marco Blanchette, Keith Joung, Anders Hansen, Sarah Johnstone, and Christine Eyler. Collaborators contributed both data and feedback on project. Corriene also presented this work at the Cold Spring Harbor Laboratory Epigenetics & Chromatin meeting in September 2022.
Randy Williams
Academic Advisor: Giovanni Parmigiani
Randy is a first year PhD student in the Biostatistics PhD program. His research interests are developing statistical methods for epi/genomic data.
During the 2022 fall term Randy is enrolled in the following courses: BIOSTAT 240: Probability II, BMI 705: Precision Medicine II, ID 201: Foundations for Public Health, BIOSTAT 232: Methods, BST 227: Introduction to Statistical Genetics, EPI 249: Molecular Biology for Epidemiologists, EPI 507: Genetic Epidemiology. And in the Spring 2023 term Randy plans to enroll in the following courses: BST 234: Introduction to Data Structures and Algorithms, BST 231: Statistical Inference I, and EPI511: Advanced Population and Medical Genetics