Text mining reveals Alzheimer's biomarkers

Searching for Alzheimer's disease biomarkers can be improved through the use of text mining, according to U.K. researchers. In a study published in the Journal of Translational Medicine, the authors were able to identify 25 biomarker candidates via the data mining of publicly available databases, and say the practice could be applied to other disorders.

"We prove that text mining works, and we will take this forward in our hunt for Alzheimer's biomarkers," lead author Simon Lovestone, PhD, a professor of old age psychiatry at King's College London's Institute of Psychiatry, said in a statement. "Our results also demonstrate the value of large data in biomedical science; you could go beyond Alzheimer's disease, and use the same approach for other conditions where biomarkers are needed, from cancer to diabetes."

The researchers developed axioms about what a blood biomarker looks like, and then used textual and linguistic analysis to develop the computer code applied to the databases. This derived a total of 25 potential biomarkers. The team then validated these and found that some had previously been identified as potential biomarkers, and in two other cases, they examined the proteins against large sample sets, and showed that the computer approach was correct.

In related news, four Johns Hopkins professors wrote an article this week on the increasing use of computational models in healthcare. Their article, published in Science Translational Medicine, concluded that such models are helping researchers identify and treat complex diseases. Also, an algorithm developed by scientists at New York City-based Mount Sinai School of Medicine is assisting in the building of networks from data found in medical records by helping researchers better understand interactions like gene-gene, protein-protein and drug side effects.