Project Overview
Artificial Intelligence and Mathematical Modeling Meet Biology
Department(s)
Computer Science
Mathematics
Biology
Abstract
My laboratory investigates the principles that govern complex biological systems and human disease by integrating approaches from mathematics, computer science, and the life sciences. We use mathematical modeling, statistical inference, machine and deep learning, natural language processing, network science, and computational analysis to address questions spanning mental health, neurodegeneration, cancer, and biological timekeeping. Since 2010, I have mentored more than 100 undergraduate researchers from biology, mathematics, computer science, neuroscience, psychology, and related disciplines. Our collaborative work has led to over 20 peer-reviewed publications, numerous presentations at national and international conferences, and a strong record of alumni placements in top graduate programs (Harvard, Columbia, UCLA, UCSD, Johns Hopkins, Carnegie Mellon, University of Chicago) as well as leading companies such as Google, Microsoft, Amazon, Meta, and Illumina.
Every summer, students join my group to work on projects at the interface of computation and biology in a supportive, intellectually active environment. I deliberately pair students with different academic backgrounds; computer science and mathematics students are not expected to know biology, and biology students are not expected to have prior programming experience: curiosity, persistence, and a genuine interest in learning matter far more than any specific skill set. Training and mentorship are built into every project.
Several projects we are interested in tackling this summer are listed below. Students are warmly encouraged to speak with me about any project. I am always happy to explain the scientific motivation, the computational or experimental approaches involved, and what day-to-day research looks like for undergraduates in my group.
Project 1: Understanding Depression and PTSD Through Language, Speech, and Facial Behavior
Although depression and PTSD share several symptoms, individuals express these conditions differently through language, vocal patterns, and facial behavior. In this project, students will analyze recorded speech, facial-expression dynamics, and linguistic content to identify behavioral signatures that differentiate the two disorders. We will apply natural language processing, machine/deep learning, and statistical analysis to uncover these patterns and examine whether they vary across sexes. This project appeals to students interested in mental health, linguistics, Artificial Intelligence, clinical research, or human behavior. It contributes to the development of more objective and accessible mental-health assessment tools.
Project 2: Predicting Parkinson’s Disease Progression using Machine/Deep Learning on Clinical and Genetic Data
Parkinson’s disease progresses at dramatically different rates across individuals. This project aims to forecast disease progression by integrating clinical measurements with genetic information. Students will analyze time-series data, construct machine/deep learning models (e.g., attention-based LSTMs), and investigate biological features associated with rapid decline. This project offers a meaningful opportunity to work at the intersection of computation, medicine, and translational neuroscience.
Project 3: Using Big Data and Artificial Intelligence to Understand the Biology of Women’s Cancers
Breast, cervical, and endometrial cancers share molecular characteristics that traditional staging systems often overlook. Using multiple molecular datasets (e.g., gene expression and other “omics” layers), students will apply machine/deep learning, network analysis, and statistical analysis to identify patterns associated with cancer stage and to discover biological pathways common across these cancers. This project is ideal for students interested in computational biology, cancer genomics, precision medicine, or large-scale data science.
Project 4: Designing Better Genes for Biotechnology using Deep Learning
The fast-growing bacterium Vibrio natriegens has enormous potential for protein production and biotechnology. In this project, students will use deep-learning models, bioinformatics algorithms, and sequence-level analysis to design DNA sequences that the organism can translate more efficiently. This computational project introduces students to Artificial Intelligence-driven biological design and has applications in synthetic biology, molecular genetics, and biotechnology.
Project 5: Discovering New Alzheimer’s Treatments with Graph-Based Neural Networks
Developing new Alzheimer’s drugs is difficult and costly, but many existing drugs may hold untapped therapeutic potential. This project uses large biomedical knowledge graphs, graph neural networks, and modern machine/deep learning tools to identify promising drug-repurposing candidates. Students will work with Python, PyTorch Geometric, network science, and interpretable models to investigate one of the most pressing challenges in biomedical research.
Project 6: Understanding How Living Systems Keep Time
Biological clocks govern essential rhythmic processes—from daily sleep–wake cycles in humans to the formation of repeating structures in developing embryos. We study two such systems: (i) the human circadian clock, which orchestrates daily physiological rhythms, and (ii) the vertebrate segmentation clock, which patterns the developing body plan. Students will use mathematical modeling, numerical simulations, optimization methods, and experimental data to explore how these clocks generate precise oscillations. This project is well-suited for students who enjoy theoretical reasoning, mathematical modeling, and fundamental questions about biological organization.
If any of these projects resonate with your interests, please reach out. I would be delighted to discuss the science, the methods, and how you might contribute.
Student Qualifications
Computer science students should ideally have taken COSC 102 and/or COSC 202, and mathematics students should have taken MATH 260 and/or MATH 376. Biology students are expected to have strong literature-search and scientific writing skills. Students who have completed BIOL 320, MATH 354, MATH 454, COSC 410, or COSC 426 are especially encouraged to apply. If you are unsure whether your background is a good match, please feel free to reach out. Students join the lab from many different starting points, and we can help you find a project where you will thrive.
Project Length
10 weeks
APPLY