Thema der Dissertation:
Performance-Driven Algorithm Engineering Optimising Pairwise Sequence Alignment and Pattern Matching Algorithms in the Era of Pangenomic Sequence Analysis Thema der Disputation:
Sequence-to-Graph Alignment
Performance-Driven Algorithm Engineering Optimising Pairwise Sequence Alignment and Pattern Matching Algorithms in the Era of Pangenomic Sequence Analysis Thema der Disputation:
Sequence-to-Graph Alignment
Abstract: Advances in sequencing technologies have led to an unprecedented growth of sequencing data as well as the compilation of databases that store the genetic diversity of entire populations of humans and other organisms. The typical pipeline used to identify genomic variants from a sequencing experiment involves four key steps: quality control, read error correction, read mapping and variant genotyping. The accuracy of the last step is highly dependent on the accuracy of the previous three steps, with this presentation focusing on reference-guided read mapping. In general, the choice of the reference sequence can significantly affect the accuracy of variant genotyping due to a phenomenon known as reference bias. A promising approach to solve this problem is to align reads against a panel of reference sequences, also known as pangenome, represented in form of a sequence graph. In this talk, I will introduce the general concept of aligning sequences to arbitrary sequence graphs and present computational methods developed for scaling sequence-to-graph alignments on whole-genome sequencing data.
Zeit & Ort
15.12.2023 | 13:00
Seminarraum 007/008
(Fachbereich Mathematik und Informatik, Arnimallee 6, 14195 Berlin)