MOLIERE: Automatic Biomedical Hypothesis Generation System
This system addresses the need for time-saving hypothesis generation in biomedical research, though it appears incremental as it builds on existing methods like Latent Dirichlet Allocation applied to network paths.
The authors tackled the problem of biomedical hypothesis generation by creating MOLIERE, a system that uses a multi-modal network from over 24.5 million documents to automatically discover connections between concepts like genes and diseases, and demonstrated its effectiveness on historical data.
Hypothesis generation is becoming a crucial time-saving technique which allows biomedical researchers to quickly discover implicit connections between important concepts. Typically, these systems operate on domain-specific fractions of public medical data. MOLIERE, in contrast, utilizes information from over 24.5 million documents. At the heart of our approach lies a multi-modal and multi-relational network of biomedical objects extracted from several heterogeneous datasets from the National Center for Biotechnology Information (NCBI). These objects include but are not limited to scientific papers, keywords, genes, proteins, diseases, and diagnoses. We model hypotheses using Latent Dirichlet Allocation applied on abstracts found near shortest paths discovered within this network, and demonstrate the effectiveness of MOLIERE by performing hypothesis generation on historical data. Our network, implementation, and resulting data are all publicly available for the broad scientific community.