AI Seminar: Accurate somatic mutation detection using weakly supervised deep learning
Join us for a talk by Anders Skanderup, Principal Investigator at the Genome Institute of Singapore.
Accurate somatic mutation detection using weakly supervised deep learning
Cancer is a genetic disease caused by mutations that lead to abnormal cell growth and behaviour. Identification of somatic mutations in tumor samples is commonly based on statistical methods in combination with heuristic filters. Here we develop VarNet, an end-to-end deep learning approach for identification of somatic variants from aligned tumor and matched normal DNA reads. VarNet is trained using image representations of 4.6 million high-confidence somatic variants annotated in 356 tumor whole genomes. We benchmark VarNet across a range of publicly available datasets, demonstrating performance often exceeding current state-of-the-art methods. Overall, our results demonstrate how a scalable deep learning approach could augment and potentially supplant human engineered features and heuristic filters in somatic variant calling.
Anders Jacobsen Skanderup is a Principal Investigator at the Genome Institute of Singapore. He also holds adjunct positions at the National Cancer Center Singapore and Dept. of Computer Science at NUS. His research focus is on using integrative and data-intensive computational approaches to decipher the genetic and molecular basis of cancer. Lab web site: http://www.skandlab.org
Anyone can participate in the event, but please use this form to register.
This seminar is a part of the AI Seminar Series organised by SCIENCE AI Centre. The series highlights advances and challenges in research within Machine Learning, Data Science, and AI. Like the AI Centre itself, the seminar series has a broad scope, covering both new methodological contributions, ground-breaking applications, and impacts on society.