Computer Science and Mathematics Meet Biology
AbstractMy lab's research focuses on using mathematical modeling, machine learning, and computer programming to understand better biological systems and human diseases. I've supervised over 70 research students from diverse backgrounds (e.g., Computer Science, Mathematics, and Biology) since 2010. More than 25 research students contributed to the publication of over 15 articles. Additionally, my students delivered research presentations at national and international conferences. More than 20 former research students now work for companies like Google, Facebook, Square, and Illumina, and more than 20 students are pursuing Ph.D., M.D., or M.S. degrees at Columbia, Harvard, Yale, UCLA, UCSD, and the University of Chicago.
This summer, we will tackle four exciting projects using an extensive array of computational techniques including machine learning, mathematical modeling, and statistical analysis. Please note that knowing the biology behind these questions is neither expected nor required.
Project 1. Prediction of Human Anxiety and Depression Disorders using Machine Learning
Mood disorders such as anxiety and depression affect a sizable portion of the population today. Several studies have discovered a link between mood disorders and misalignments in human circadian rhythms caused by variations in circadian clock genes. This study will use machine learning and statistical analysis to identify the most critical genetic variations and clinical factors for mood disorders and then use these risk factors to build a predictive model for mood disorders.
Project 2. Predicting Human Cancers from the Gut Microbiome: A Machine Learning Approach
The number of microbiome-related studies has recently increased. These studies contribute to our understanding of the host-microbiome relationship and how it influences the development and progression of complex diseases. Establishing a link between the microbiome and disease states is advantageous for personalized medicine. Using machine learning and statistical analysis, this project will investigate the role of the human microbiome in patients’ disease status and how microbiome data can be used for personalized medicine.
Project 3. Can Machine Learning Predict Malnutrition in Children?
Malnutrition (e.g., undernutrition and vitamin deficiency) has long-term developmental, economic, social, and medical consequences for individuals, families, communities, and countries. For example, undernutrition accounts for approximately 45% of infant deaths in the world. This project will use a detailed Ethiopian school children survey to identify the potential risk factors for malnutrition. Using these risk factors, we will develop machine learning approaches to predict the malnutrition status of kids.
Project 4. Developing a new GPU Based Software for Simulating Large Scale Biological Systems
There is no stochastic simulation software for large-scale biological systems with delays. This project will create a high-performance parallel version of an existing stochastic simulation software. The new software should create highly compact and efficient data structures and potentially support GPU acceleration.
Student QualificationsStudents must have taken COSC 102, MATH 260, and/or MATH 376.
Number of Student Researchers4 students
Project Length10 weeks
Applications open on 01/03/2022 and close on 02/04/2022