NetMe 2.0: a web-based platform for extracting and modeling knowledge from biomedical literature as a labeled graph

Abstract

Motivation: The rapid increase of bio-medical literature makes it harder and harder for scientists to keep pace with the discoveries on which they build their studies. Therefore, computational tools have become more widespread, among which network analysis plays a crucial role in several life-science contexts. Nevertheless, building correct and complete networks about some user-defined biomedical topics on top of the available literature is still challenging.

Results: We introduce NetMe 2.0, a web-based platform that automatically extracts relevant biomedical entities and their relations from a set of input texts-i.e. in the form of full-text or abstract of PubMed Central's papers, free texts, or PDFs uploaded by users-and models them as a BioMedical Knowledge Graph (BKG). NetMe 2.0 also implements an innovative Retrieval Augmented Generation module (Graph-RAG) that works on top of the relationships modeled by the BKG and allows the distilling of well-formed sentences that explain their content. The experimental results show that NetMe 2.0 can infer comprehensive and reliable biological networks with significant Precision-Recall metrics when compared to state-of-the-art approaches.

Availability and implementation: https://netme.click/.

Figure 1.

(a) Dependency-parse tree of the sentences: “TP53 expression increased in colon cancer.” (b) The three mentions “cell viability,” “cell motility,” and “circPIP5K1A overexpression” have not been annotated by OntoTagMe, thus the three verbs “attenuates, reduces, facilitates” are used to annotate the relationship between the detected mentions “circPIP5K1A” and “colon cancer.” (c) On-the-fly Graph-RAG approach. Users send biomedical queries (1) with the NetMe 2.0 GUI. Next, the knowledge graph is generated (2) by analyzing a collection of documents (from PubMed or PubMed Central) related to the user query and visualized via the GUI (3). Then, the user can select a set of nodes of interest (4), which are passed to the Sentences Retrieval module (5) to extract some phrases associated with the paths connecting such entities (6). These sentences are then transmitted to OpenAI (7) to generate a summarized text (8) explaining the (biomedical) relationships among those entities.