Congratulations to Xuan for the acceptance of the “pathway enrichment” paper. In this research, Xuan developed a set of new metrics to evaluate the effective of pathway enrichment in terms of text mined vital gene set. This research reflect an interesting view, from which BioNLPers could observe how the literature-mined entities contribute to the downstream bioinformatics task.
Abstract goes here:
Background: Natural language processing has long been applied in various applications for biomedical knowledge inference and discovery. Enrichment analysis based on named entity recognition is a classic application to infer enriched associations in terms of specific biomedical entities like like gene, chemical, mutation, and so on.
Objective: To investigate the effect of enrichment evaluation with respect to Biomedical text mining in pathway enrichment, a series of novel metrics, Inverse Pathway Frequency (IPF), are proposed in this research. To test the robustness of the proposed metrics, IPF metrics and traditional evaluation metric are compared in simulation experiments.
Methods: First, a couple of biomedical text mining strategies are applied to investigate drug-related genes set. Second, novel metrics for pathway enrichment of text mined genes are designed in the support of drug repurposing. In this case, seven novel IPF metrics are proposed and compared with traditional p-value metric.
Results: IPF metrics are evaluated in a case study of rapamycin-related gene set. By applying the best IPF metric in a pathway enrichment simulation test, a novel discovery of drug efficacy of rapamycin for breast cancer is replicated from the chosen data prior to year 2000. This result shows positive effect of the best IPF metric in the support of new knowledge discovery in drug repurposing. Furthermore, the mechanism hidden in the drug-disease association is visualized by Cytoscape.
Conclusions: The results suggest the effectiveness of the proposed IPF metrics in pathway enrichment evaluation, as well as the application in drug repurposing.