The Davis lab is currently looking for an enthusiastic undergraduate with to join the team and work on a computational project in the lab - see the posting below and feel free to contact me directly at email@example.com if you're interested:
Project Title: Accelerating proteomics data analysis using highly parallel GPU-based processing
Project Description: High resolution mass analysis of proteins and peptides reveals complex spectra that must be accurately fit to determine the relative abundance of analytes. This fitting process is computationally intensive, but is readily amenable to parallelization using modern graphic processing units (GPUs). In this project, we will implement a least-squares Fourier transform convolution algorithm in CUDA and apply this algorithm to the analysis of a variety of mass spectra generated in our laboratory. The project includes the following aspects:
- Implementation of a fitting algorithm on GPU hardware
- Use of a top-of-the-line mass spectrometer to generate reference mass spectra for analysis
- Application of this fitting algorithm to the quantitation of spectra generated above
- Analysis of the benefits of GPU-based processing over traditional CPU-based implementations.
This UROP is offered in the Biology department, however highly motivated applicants from departments such as Electrical Engineering & Computer Science, Mathematics, and Biological Engineering are encouraged to apply. While this UROP is primarily computational, students interested in learning more about mass spectrometry will have significant opportunities to work in the wet laboratory. This UROP project will be directly supervised by Prof. Joey Davis.
Prerequisites: Candidates of all experience levels will be considered. However, preference is given to candidates with experience with CUDA and/or python who can commit to working at least 20 hours per week during IAP and summer and 9 hours per week during the academic year. I am offering academic credit for new UROPs.
This project is expected to require IAP/Spring terms for completion, however the UROP can be expanded to include additional projects in the coming years.