Development of a Drug Discovery Informatics System
The Initiative 'Development of a Drug Discovery Informatics System' was conducted for five years from 2015 to support drug discovery and development in Japan. In this project, in order to improve the pharmacokinetics and toxicity which was bottleneck in academia drug discovery, in silico prediction systems of drug metabolism and pharmacokinetics, cardiotoxicity and hepatotoxicity were constructed. These in silico prediction systems are expected to be a practical tool for prioritization and optimization of compound synthesis in the early stage of drug discovery.
DruMAP is an analysis platform for various pharmacokinetic parameters that are important for drug development. DruMAP consists of a database of curated public data and experimental data obtained under unified conditions and programs which predicts pharmacokinetic parameters only from chemical structures using machine learning techniques. Users can predict those parameters for novel compounds.
CypSOM predictor is a SOM (site of metabolism in a drug molecule) prediction tool by using the molecular simulation techniques (molecular dynamics simulation, molecular docking and SMARTCyp). The SOM prediction score is estimated from the accessibility score and the reactivity score based on the 3D structures of the CYP protein and the drug molecule.
AMED Cardiotoxicity Database is a database of small molecules which bind to various ion
channels and potentially cause cardiotoxic risk.
AMED Cardiotoxicity Database compiles cardiotoxicity-related information from publicly available databases and integrates them in standardized format. As an initial target, bioactivities for hERG potassium channel were collected from ChEMBL, NIH Chemical Genomics Center, and hERGCentral because the inhibition of hERG potassium channel is closely related to the prolonged QT interval, and to assess the risk could greatly contribute to avoid delay of the development of therapeutic compounds or withdrawal of marketed drugs.
DILI-TOOLBOX represents the portal site for all databases and prediction systems developed in the AMED Liver Toxicity Informatics System Development programs.
DILI-cSEARCH is a freely available database that provides comprehensive information about liver toxicity caused by drugs and chemicals. This database offers cross search access to information of Open TG-GATEs, DrugMatrix, LINCS and LTKB (Liver Toxicity Knowledge Base) including in vitro/in vivo experimental conditions, toxicological profiles, gene expression profiles, and much more. Available data may be searched in various ways.
DILI-PANEL is a prediction system of liver toxicity that was established to support the safety assessment and risk management for drug candidates. This system can automatically derive toxicological alerts from gene expression data of in vitro human hepatocytes using marker panel of liver toxicity. The marker panel of liver toxicity was established using the gene expression data of human primary hepatocytes in DILI-cSEARCH and machine learning methods.
TOXPILOT is a toxic process interpretable knowledge system. Focusing on hepatotoxicity, TOXPILOT provides visualization maps of the toxic course, general course map that visualizes general toxic courses common to multiple specific toxic courses, and systematized knowledge for hepatotoxic processes based on Toxic Process Ontology, TXPO.
As part of QSAR, LIVER/MIE-QSAR system has served as predictor for 37 types of activities of molecular initiating events (MIE). Within the system, MIE activities such as nuclear receptors and stress response pathways are predictable through machine learning. The prediction effectively deal with the toxicity of drug-induced liver injury (DILI) based on the computation of their chemical structure.
As part of QSAR, LUNG/MIE-QSAR system has served as predictor for 37 types of activities of molecular initiating events (MIE). Within the system, MIE activities such as nuclear receptors and stress response pathways are predictable through machine learning. The prediction effectively deal with the toxicity of drug-induced pulmonary toxicity (DIPT) based on the computation of their chemical structure.
toxBridge is a database that stores statistical analysis results using differential gene expression when compounds are exposed. It is possible to display feasible bridging capabilities between rat in vivo/vitro and human in vitro based on signature genes and pathways.