Knowledge integration and decision support for accelerated discovery of antibiotic resistance genes

Abstract

We present a machine learning framework to automate knowledge discovery through knowledge graph construction, inconsistency resolution, and iterative link prediction. By incorporating knowledge from 10 publicly available sources, we construct an Escherichia coli antibiotic resistance knowledge graph with 651,758 triples from 23 triple types after resolving 236 sets of inconsistencies. Iteratively applying link prediction to this graph and wet-lab validation of the generated hypotheses reveal 15 antibiotic resistant E. coli genes, with 6 of them never associated with antibiotic resistance for any microbe. Iterative link prediction leads to a performance improvement and more findings. The probability of positive findings highly correlates with experimentally validated findings (R2 = 0.94). We also identify 5 homologs in Salmonella enterica that are all validated to confer resistance to antibiotics. This work demonstrates how evidence-driven decisions are a step toward automating knowledge discovery with high confidence and accelerated pace, thereby substituting traditional time-consuming and expensive methods.

KGLM: Integrating Knowledge Graph Structure in Language Models for Link Prediction

“…The model performed equally well as the state-of-the-art in cases where the training dataset had only ontological relationships, such as the /music/artist/origin relation present in the FB15k-237 dataset. In future work, we plan to further test the version of KG LM that takes into account entity types, KGLM GER , on domain-specific knowledge graphs like KIDS [6] with entity types in their schema, for example, encoding that the entity "ampicillin" is of type "antibiotic". One additional improvement would be to adopt Siamesestyle textual encoders [18], [39], which can considerably decrease the training and inference times.…”

“…It is represented with a set of triples, where a triple consists of (head entity, relation, tail entity) or (h, r, t) for short, for example (Bill Gates, founderOf, Microsoft) as shown in Figure 1. Due to their effectiveness in identifying patterns among data and gaining insights into the mechanisms of action, associations, and testable hypotheses [2], [3], both manually curated KGs like DBpedia [4], WordNet [5], KIDS [6], and CARD [7], and automatically curated ones like FreeBase [8], Knowledge Vault [9], and NELL [10] exist. However, these KGs often suffer from incompleteness.…”

From Data to Wisdom: Biomedical Knowledge Graphs for Real-World Data Insights

“…When data and information are joined in such a structure, the resulting data representation is referred to as a knowledge graph , which provides a computationally accessible (i.e., machine-readable) representation of relationships between disparate biologic systems information. From these KGs, novel relationships can be identified and used to generate wisdom and actionable insights, as demonstrated in recent publications from the fields of computational biology and the life sciences [ 7 - 9 ]. In Fig.…”

Fast, lightweight, and accurate metagenomic functional profiling using FracMinHash sketches

“…This integration represents an exciting avenue for advancing our comprehension of intricate biological systems: metagenomic functional profiles provide insights into the functional potential of microbial communities; KEGG's rich resource of hierarchical modules and pathways offers a structured framework to interpret and contextualize these functions; and knowledge graphs, which link genes, proteins, diseases, drugs, and other biological entities, can facilitate the discovery of underlying relationships between different entities. By connecting functional profiles with biomedical knowledge graphs, researchers can explore the functional intersections of diseases [55,49], identify potential therapeutic targets [75,78,46], and predict novel associations between microbial functions and human health [15,43]. Pursuing this integrated approach can not only aid in hypothesis generation but also pave the way for innovative research.…”