Exploiting semantic patterns over biomedical knowledge graphs for predicting treatment and causative relations

Abstract

Background Identifying new potential treatment options for medical conditions that cause human disease burden is a central task of biomedical research. Since all candidate drugs cannot be tested with animal and clinical trials, in vitro approaches are first attempted to identify promising candidates. Likewise, identifying different causal relations between biomedical entities is also critical to understand biomedical processes. Generally, natural language processing (NLP) and machine learning are used to predict specific relations between any given pair of entities using the distant supervision approach. Objective To build high accuracy supervised predictive models to predict previously unknown treatment and causative relations between biomedical entities based only on semantic graph pattern features extracted from biomedical knowledge graphs. Methods We used 7,000 treats and 2,918 causes hand-curated relations from the UMLS Metathesaurus to train and test our models. Our graph pattern features are extracted from simple paths connecting biomedical entities in the SemMedDB graph (based on the well-known SemMedDB database made available by the U.S. National Library of Medicine). Using these graph patterns connecting biomedical entities as features of logistic regression and decision tree models, we computed mean performance measures (precision, recall, F-score) over 100 distinct 80%–20% train-test splits of the datasets. For all experiments, we used a positive:negative class imbalance of 1:10 in the test set to model relatively more realistic scenarios. Results Our models predict treats and causes relations with high F-scores of 99% and 90% respectively. Logistic regression model coefficients also help us identify highly discriminative patterns that have an intuitive interpretation. We are also able to predict some new plausible relations based on false positives that our models scored highly based on our collaborations with two physician co-authors. Finally, our decision tree models are able to retrieve over 50% of treatment relations from a recently created external dataset. Conclusions We employed semantic graph patterns connecting pairs of candidate entities in a knowledge graph as features to predict treatment/causative relations between them. We provide what we believe is the first evidence in direct prediction of biomedical relations based on graph features. Our work complements lexical pattern based approaches in that the graph patterns can be used as additional features for weakly supervised relation prediction.

<p>Integrating Unified Medical Language System and Kleinberg’s Burst Detection Algorithm into Research Topics of Medications for Post-Traumatic Stress Disorder</p>

“…However, with further research, when the frequency of burst words reaches the threshold of high-frequency words, the negative effect may be gradually discovered and reported. Therefore, the next step is to integrate other models 52 to mine some plausible, false positive or negative semantic relations.…”

Drug-Drug Interaction Prediction on a Biomedical Literature Knowledge Graph

“…Such an example is the method proposed in [5], which builds a heterogeneous network and performs link prediction to construct an integrative model of drug efficacy. Most relevant to our approach is the work in [2,17], presenting drug discovery methods, based on biomedical knowledge graphs. The former method focuses on treatment and causative relations exploiting connections of biomedical entities as found in literature.…”

A Framework for Automated Construction of Heterogeneous Large-Scale Biomedical Knowledge Graphs

“…To date, the vast majority of KG construction algorithms have been developed in order to create more manageable representations of large free-text corpora (e.g. scientific articles) [9,10] , to derive novel associations between existing concepts [11,12] , and add evidence to existing systems or KGs [13,14] . While many data-driven KG construction methods have been developed, they remain largely unable to automatically construct KGs from multiple disparate data sources, combine KGs created by different systems, and collaborate or share KGs across institutions due to their inability to account for the use of different schemas, standards, and vocabularies [15] .…”