Computers learn to spot deadly food poisoning bacteria

Most E. coli bacteria live in cattle and are harmless to them but some strains can spread to humans and cause deadly disease. Until now, telling which ones are likely to be able to do this has been very difficult. A study, led by the University of Edinburgh’s Roslin Institute and published in the Proceedings of the National Academy of Sciences, has shown that data analytics and artificial intelligence can predict which bacterial strains commonly found in cattle could spread to people. The scientists used genetic information specific to 185 different E. coli strains to train an algorithm, using machine learning, to tell whether they came from human or bovine hosts. The researchers then tested how the computer would classify DNA from two new E. coli food poisoning outbreaks related…


Link to Full Article: Computers learn to spot deadly food poisoning bacteria