A research team led by Professor Ruibang LUO at HKU's School of Computing and Data Science developed ClairS, a deep-learning algorithm that improves cancer mutation detection using long-read sequencing. Tested on breast cancer, lung cancer and melanoma cell-line datasets, ClairS demonstrated exceptional accuracy across various cancer types and sequencing conditions. It addresses limitations of short-read sequencing methods in structurally complex genome regions by utilizing long-read sequencing. The team designed a strategy mixing sequencing data from normal human samples to generate synthetic tumour-normal data, allowing ClairS to create unlimited realistic training examples. ClairS has been tested on multiple cancer cell lines including breast, lung, melanoma and pancreatic cancers, showing high accuracy in detecting small cancer mutations. It has been integrated into Oxford Nanopore Technologies’ official somatic variant-calling workflow. The research was published in Nature Methods titled “ClairS: a deep-learning method for long-read tumor–normal pair somatic small variant calling”. ClairS is open source and available at https://github.com/HKU-BAL/ClairS.