Researchers will use a prototype of IBM's cloud computing system specifically adapted for genomic research to read and interpret vast amounts of genetic data to help tailor treatment for a common type of brain cancer.
The project is starting with 25 patients affected by glioblastoma multiforme, an aggressive and malignant brain cancer that is usually fatal within a few months. However, it usually takes longer than this to sequence a patient and their tumour's respective genomes and then to use this information to discover the best targets for treatment. Watson is expected to reduce the time it takes to correlate a patient's genetic data with other relevant literature.
Dr Robert Darnell, New York Genome Center's president and scientific director, explains that analysing a patient's brain tumour involves sequencing around 800 billion DNA base pairs. The DNA data must then be combined with a patient's clinical data, medical literature, study findings and other relevant material to enable doctors to consider which of a variety of treatment options are best suited to a patient.
Once researchers have sequenced the patient's genome and that of the tumour using rapid genome sequencing, they feed all the data into Watson. Because the system uses artificial intelligence, it can 'learn' what proteins are produced by the cancer by scouring through huge volumes of clinical research for an answer and then apply this knowledge to subsequent problems or treatment strategies.
'As genomic research progresses and information becomes more available, we aim to make the process of analysis much more practical and accessible through cloud-based, cognitive innovations like Watson', said Dr John Kelly, research director at IBM. He told BBC News: 'It's like big data on steroids'.
'Watson can do in seconds what would take people years. And we can get it down to a really personal level. This is the proverbial needle in the haystack and the haystack is enormous'.