Literature-based automated discovery of tumor suppressor p53 phosphorylation and inhibition by NEK2.

摘要: Scientific progress depends on formulating testable hypotheses informed by the literature. In many domains, however, this model is strained because number of research papers exceeds human readability. Here, we developed computational assistance to analyze biomedical literature reading PubMed abstracts suggest new hypotheses. The approach was tested experimentally tumor suppressor p53 ranking its most likely kinases, based all available abstracts. Many best-ranked kinases were found bind and phosphorylate (P value = 0.005), suggesting six so far. One these, NEK2, studied in detail. A known mitosis promoter, NEK2 shown at Ser315 vitro vivo functionally inhibit p53. These bona fide validations text-based predictions phosphorylation, discovery an inhibitory kinase pharmaceutical interest, that automated reasoning using a large body can generate valuable molecular has potential accelerate scientific discovery.