A mutation-centric approach to identifying pharmacogenomic relations in text

► Identifying genetic variants is crucial to extracting pharmacogenomic relations. ► Identifying genetic variants characterizes our mutation-centric approach. ► Our approach achieves a recall of 33-46% and a precision of 65%. ► Our performance is roughly similar to Stanford's relation-centric approach. ► We show that mutation- and relation-centric approaches are complementary.