Integrating Mendelian randomization and literature-mined evidence for breast cancer risk factors (preprint)

An increasing challenge in population health research is efficiently utilising the wealth of data available from multiple sources to investigate the mechanisms of disease and identify potential intervention targets. The use of biomedical data integration platforms can facilitate evidence triangulation from these different sources, improving confidence in causal relationships of interest. In this work, we aimed to integrate Mendelian randomization (MR) and literature-mined evidence from the EpiGraphDB knowledge graph to build a comprehensive overview of risk factors for developing breast cancer. We utilised MR-EvE (“Everything-vs-Everything”) data to generate a list of causal risk factors for breast cancer, integrated this data with literature-mined relationships and identified potential mediators. We used multivariable MR to evaluate mediation and estimate the direct effects of these traits. We identified 213 novel and established lifestyle and molecular traits with evidence of an effect on breast cancer. We present the results of this evidence integration for four case studies (insulin-like growth factor I, cardiotrophin-1, childhood body size and age at menopause). We demonstrate that using MR-EvE to identify disease risk factors is an efficient hypothesis-generating approach. Moreover, we show that integrating MR evidence with literature-mined data may identify causal intermediates and uncover the mechanisms behind disease.

Triangulating evidence in health sciences with Annotated Semantic Queries (preprint)

ABSTRACT Integrating information from data sources representing different study designs has the potential to strengthen evidence in population health research. However, this concept of evidence “triangulation” presents a number of challenges for systematically identifying and integrating relevant information. We present ASQ (Annotated Semantic Queries), a natural language query interface to the integrated biomedical entities and epidemiological evidence in EpiGraphDB, which enables users to extract “claims” from a piece of unstructured text, and then investigate the evidence that could either support, contradict the claims, or offer additional information to the query. This approach has the potential to support the rapid review of pre-prints, grant applications, conference abstracts and articles submitted for peer review. ASQ implements strategies to harmonize biomedical entities in different taxonomies and evidence from different sources, to facilitate evidence triangulation and interpretation. ASQ is openly available at https://asq.epigraphdb.org .

Multi-omics study revealing tissue-dependent putative mechanisms of SARS-CoV-2 drug targets on viral infections and complex diseases (preprint)

Drug target prioritisation for new targets and drug repurposing of existing drugs for COVID-19 treatment are urgently needed for the current pandemic. Here we pooled 353 candidate drug targets of COVID-19 from clinical trial registries and the literature and estimated their putative causal effects in 11 SARS-CoV-2 related tissues on 622 complex human diseases. By constructing a disease atlas of drug targets for COVID-19, we prioritise 726 target-disease associations as evidence of causality using robust Mendelian randomization (MR) and colocalization evidence (http://epigraphdb.org/covid-19/ctda/). Triangulating these MR findings with historic drug trial information and the druggable genome, we ranked and prioritised three genes DHODH, ITGB5 and JAK2 targeted by three marketed drugs (Leflunomide, Cilengitide and Baricitinib) which may have repurposing potential with careful risk assessment. This study evidences the value of our integrative approach in prioritising and repurposing drug targets, which will be particularly applicable when genetic association studies of COVID-19 are available in the near future.