HKU Develops ClairS for Enhanced Cancer Mutation Detection Across Multiple Types
A research team at the University of Hong Kong, led by Professor Ruibang Luo, has developed ClairS, a deep-learning algorithm designed to improve the detection of cancer mutations using long-read sequencing. ClairS has been tested on datasets from breast cancer, lung cancer, and melanoma, demonstrating high accuracy across various cancer types. The algorithm addresses challenges faced by existing methods that rely on short-read sequencing, which often struggle with complex genomic regions. ClairS uses a novel strategy to generate synthetic tumor-normal data, allowing for the creation of numerous realistic training examples. This innovation supports the development of a robust AI model for cancer genomic analysis. ClairS has been integrated into Oxford Nanopore Technologies' somatic variant-calling workflow, marking a significant step towards the application of advanced sequencing technologies in clinical genomics. 
