(External) Modern Techniques in Sequence Alignments

Advisor: Pauline Cheung, Luminex Corporation

Proposed UoG coadvisors: Leonarda Susta, Pathobiology; Baozhong Meng, Molecular and Cellular Biology; Stefan Kremer, Computer Science

Traditional sequence alignment strategies rely on recursive pairwise comparisons of sequences followed by adjustments. These algorithms have several flaws, including exponential scaling of computational time with the number of sequences and the generation of erroneous gaps in the multi-sequence alignment. With the recent advent of AI and deep learning technologies, better strategies have cropped up that potentially offer better solutions. In the proposed project, the student will first conduct a survey of the new AI and deep learning technologies developed for multi-sequence alignments. The student will then apply one or more of these technologies in the alignment of the family of enteroviruses and rhinoviruses and compare the performance to traditional algorithms.

This is a one-semester, remote project.