Project Mentor

Dr. Pedro Mendes 
Center for Quantitative Medicine, Center for Cell Analysis and Modeling (Department of Cell Biology)

Undergraduate Research Opportunity Description

Project Description	A critical aspect in developing computational systems biology models is to estimate values for the parameters of a model based on experimental data. Our systems biology software COPASI (http://copasi.org) is one of the leading packages for parameter estimation, which is widely used in the literature (around 100 papers per year use it). However COPASI executes parameter estimation using optimization algorithms that run in serial mode and tcan be very slow. We aim to address this problem by implementing optimization algorithms, known as “island evolutionary algorithms”, that can run in parallel making use of high-performance computing resources. This project will implement such an algorithm, to be written as a script in the R programming language and which will control COPASI through an existing API (https://github.com/jpahle/CoRC). This research project includes coding, debugging and benchmarking the algorithm using established test case problems. Finally we will apply it in an ongoing research project on genetic regulation by micro-RNAs.
Project Direction	Several applications of parameter estimation are currently limited by the speed of finding paramater values for large models. Parallel algorithms that use HPC facilities (as the one at UConn Health) will be particularly useful in projects that develop models from large data sets, such as provided by genomics. But once solutions are found, it also becomes important to assess the quality of those models by approaches such as sensitivity and identifiability analyses. Current methods to process these also run in serial mode and can benefit from parallelization and would be a natural continuation of the project. We envision publication of these methods in journals such as Bioinformatics, PLOS Computational Biology, or Journal of Open Source Research.
Mentorship and Supervision	The student will be provided training in the modeling process in computational systems biology, and will be introduced to the COPASI software and its applications. There will be daily planning and progress update meetings in the first two weeks, which will then become twice weekly as the student becomes more familiar with it. The supervisor, as well as postdoctoral fellows, will be available daily to help with difficulties the student may come across in the project. Our research environment encourages frequent informal discussions with the supervisor and other members of the research team, as needed. The student will be expected to write a project report at the end of the Summer period, and to present their work in a short seminar (10min) in a session together with other Summer research students.
Student Qualifications	Programming experience required (preferably in R, but not necessary). Previous or current coursework covering topics in optimization, parallel computing, or nonlinear dynamics, would be advantageous. Ideal project for students from Computer Science, Statistics or Mathematics to expand to Computational Biology. Can also be suited to biology students who desire to gain experience in Computational Biology, though must already have some experience in programming.
Summer Schedule Options	Research Dates: May 26 to July 31, 2020 Schedule: M-F, 9am-5pm
Project Continuation	Fall 2020, Spring 2021
Academic Year Time Commitment	6  hours/week
Possible Thesis Project	Yes

Application

Submit an online application for this research opportunity at https://quest.uconn.edu/prog/HRP20-19. The application deadline is Monday, February 3, 2020.

This application requires a resume or CV, an unofficial transcript, and a brief statement of research interests.