Bethany Percha
Biomedical Informatics
Bio-X SIGF Fellow 2013/2014  
Profs. Russ Altman (Bioengineering), Chris Potts (Linguistics), and Daniel Rubin (Radiology)

Unstructured, text-based resources such as the scientific and patent literatures and electronic medical records provide a wealth of valuable biomedical data that we have barely begun to tap. Bethany's research goal is to apply modern natural language processing (NLP) and text mining methods to important biomedical problems. Specifically, she is using a combination of NLP and crowdsourcing to extract structured relationships among drugs, genes, and phenotypes (such as diseases and side effects) from unstructured text. Networks of these structured “facts” can help us better understand important biomedical phenomena like drug-drug interactions and patient-level variation in drug response.