Have a UROP opening you would like to submit?
Please fill out the form.
Machine Learning for Metabolomics [remote]
IAP and Spring
20: Biological Engineering
Our lab develops computational and experimental approaches to understand human biology. New experimental methods make it possible to measure cellular changes across the genome, epigenome, proteome, and metabolome. These technologies include genome-wide measurements of transcription, of protein-DNA interactions, of chromatin accessibility, of genetic interactions, and of protein and metabolite interactions. Each data source provides a very narrow view of the cellular changes. By computationally integrating these data we can reconstruct signaling pathways and identify previously unrecognized regulatory mechanisms. This project provides a unique opportunity for a student to join the development of a comprehensive set of machine learning tools to model biological measurements using deep neural networks and graph representation learning. You will work with a team of interdisciplinary researchers to identify relevant algorithms, build and test existing models as well as develop novel architectures to improve the current state-of-the-art.
Interest in computational biology, experience in machine learning and facility with munging large datasets; understanding of core machine learning concepts, preferred experience with deep learning platforms such as tensorflow and pytorch.