SBIR TOPIC 136 - INSILICOM: Enhancing a large-scale biomedical knowledge graph for infectious- and immune-mediated disease research

The exponential growth of biomedical literature has posed significant challenges for researchers striving to stay abreast with the vast swathes of newly published data. This surge, coupled with the enormous data generated by high-throughput technologies, underscores the need for a radical evolution in data harnessing techniques. Our proposed project addresses this challenge by advancing automated knowledge discovery (AKD), converting unstructured literature data into structured formats, and facilitating efficient and accurate information retrieval. Leveraging the cutting-edge models from our recent successes in the LitCoin NLP Challenge and BioCreative Challenges, we have constructed a large-scale biomedical knowledge graph, BioKG, by processing all PubMed abstracts up to May 2022 and integrating relation data from 40 public databases. In this project we will enhance BioKG by adding relations extracted from the latest PubMed abstracts and PMC full-text articles and implementing weekly update function. We will also improve the ontology related to infectious- and immune-mediated diseases to facilitate more effective knowledge discover applications related to these diseases. With our commitment to open-source principles, we aim to catalyze collaborative research, making complex data accessible and interpretable and guiding future scientific endeavors in infectious- and immune-mediated diseases.