Detecting a mutation that eventually leads to a cancerous tumour is as difficult as it sounds. Current technologies can combine computational tools and human expertise to sequence and analyse DNA taken from a tumour to reveal what types of genetic changes and tumour mutation have caused uncontrolled cancer growth.
However, researchers from John Hopkins University believe these methods fall short and have developed a novel tool. The Cerebro technique involves machine learning that employs a class of algorithms examining large sets of data in order to determine whether a mutation in the tumour’s DNA is causative to its growth or spread.
In studies carried out examining lung and melanoma patient samples, Cerebro was found to be more accurate when compared to six other existing tumour mutation identification tools, thus showing great promise in improving patient outcomes.
This technology development is one of many data analytics and machine learning research programs connected to the Human Cell Atlas and The Cancer Genome Atlas (TCGA).
Image: Melanoma cell by Nephron / CC-BY-SA 3.0
This story is taken from the 14 August 2018 edition of The Warren Centre’s Prototype newsletter. Sign up for the Prototype here.