31 July 2017
ByAppeared in BioNews 911
The IBM Watson supercomputer can analyse 160 hours' worth of genomic data in just ten minutes.
In a proof-of-concept study, researchers at New York Genome Centre have demonstrated this speed of analysis could prove life-saving in finding drug targets and even clinical trials for some fast-progressing cancers.
'This study is an important step forward promoting human-machine interface as a way to address a key bottleneck in cancer genomics,' claim the authors.
In the study, published in Neurology Genetics, researchers used data obtained from a patient who had died of glioblastoma – an aggressive form of brain cancer. DNA and RNA had been extracted from the tumour tissue, in addition to the patient's healthy DNA extracted from normal blood. These samples had been sequenced for comparison.
The researchers compared the supercomputer's ability to interpret whole genome sequencing (WGS) data with the work of a team of experts, which had taken them over six days. Through the use of artificial intelligence (see BioNews 747), the beta version of the automated system IBM Watson Genomic Analytics rapidly analysed this sequenced genomic data and cross-referenced findings with biomedical literature on PubMed in ten minutes.
The results quickly revealed gene mutations that could have acted as drug targets, and potential therapeutic options for the patient. The supercomputer was also able to identify a potentially relevant clinical trial.
'We identified a potentially relevant National Institutes of Health-sponsored clinical trial with targeted therapies that matched the mutations/pathways seen with WGS and RNA sequencing that were not otherwise visible,' Principal investigator Dr Robert Darnell of the The Rockerfeller University, told Medscape Medical News. While this trial had been identified and recommended to the patient, the tumour growth had progressed too rapidly for him to be eligible.
'Time was a big variable in [the patient 's] outcome... This case therefore illustrates the potential and the challenge of where high-end cancer genomics is at and where it is headed,' said Dr Darnell.
'The study documents the strong potential of Watson for Genomics to help clinicians scale precision oncology more broadly,' commented Vanessa Michelini, Watson for Genomics Innovation Leader at IBM Watson Health. 'Clinical and research leaders in cancer genomics are making tremendous progress towards bringing precision medicine to cancer patients, but genomic data interpretation is a significant obstacle, and that's where Watson can help.'