A Novel Measure For Finding Disease-Specific Genes From The Biomedical Literature

We present a novel measure that identifies disease-associated genes from the biomedical literature in terms of causing less side-effects. Our aim is to extract genes that not only have strong associations with a given disease, but also have no or weak associations with other diseases. This enables identification of specific disease-associated genes, the decreased expression of which would result in a lower probability of side-effects, thus contributing to efficient drug development. Our proposed method incorporates transitive associations between the disease and genes based on the frequency of co-occurrence of gene terms.