CompoundDB4j: Integrated Drug Resource of Heterogeneous Chemical Databases

Abstract

Computational approaches to analyze various drug/ compound centered analysis often present a need to map attributes from multiple drug databases. In this study, we provide a Neo4j repository that integrates two of the most prominent open source drug databases, DrugBank and ChEMBL, with a goal of establishing an integrated data visualization and analysis tool for drug discovery studies. The drugs present in DrugBank are mapped to their counterparts in ChEMBL. The integration of these resources and the harmonization using knowledge graph serialization using Neo4j lead to identification of relationships between drugs and other related features that are otherwise spread across two different resources. A common data format, a prerequisite to populate the Neo4j database, enables users to identify new relationships central to drug discovery research, like Drug Target Interactions (DTI). The resource is freely available at: https://github.com/ambf0632/CompoundDB4j

Predicting adverse drug reactions of two‐drug combinations using structural and transcriptomic drug representations to train an artificial neural network

“…Integrating datasets, even for common biological or chemical entities across highly curated datasets is an intensive, F I G U R E 3 Adverse Drug Reactions predicted by artificial neural network for the combination of naproxen and goserelin [Colour figure can be viewed at wileyonlinelibrary.com] time-consuming exercise (Murali et al, 2020). The scope of this project only covers the inclusion of ADRs resulting from DDIs reported in the TWOSIDES data table.…”

“…Considering the availability of various high performance python libraries, this study can be easily extended to multiple deep learning and AI algorithms with the provided data sets. Integrating datasets, even for common biological or chemical entities across highly curated datasets is an intensive, time-consuming exercise (Murali et al, 2020) . The scope of this project only covers the inclusion of ADRs resulting from DDIs reported in the TWOSIDES data table.…”

Predicting clinical trial outcomes using drug bioactivities through graph database integration and machine learning

“…Consequently, each research objective has to have its data integration strategy that is focused on serving the needs of that objective, and universal integration of all databases will not work in most ML-based biomedical research. In this context, we have previously synthesized the initial framework, CompoundDB4j (Murali et al, 2020), which contains integrated data of ChEMBL and DrugBank, to have all approved drugs in DrugBank accurately mapped to ChEMBL, such that all related annotations (e.g., bioactivities) of these compounds can be retrieved. The extended capabilities in the current instance of the framework include clinical trials (Tasneem et al, 2012) and experimentally validated disease-target associations (Davis et al, 2008) to predict clinical outcomes.…”

“…Figure 1 serves as a functional diagram to explain the integration. However, the actual integration is done as an extension to CompoundDB4j (Murali et al, 2020), which already integrates ChEMBL and DrugBank in a Neo4j (Neo4j) database. Graph databases enable the creation of complex queries that involve many connections, as compared to relational databases.…”

Combined Graph/Relational Database Management System for Calculated Chemical Reaction Pathway Data

“…In this case, represent of the ensemble of data as a graph looks more convenient, because it allows to apply various graph theory algorithms (e.g., shortest path search) for the analysis of pathways. Recently, application of a graph database architecture for chemical data has been reported 18 as a tool for merging data from ChEMBL abd DrugBank.…”