JRC QSAR Model Database
The JRC QSAR Model Database is a freely accessible web application that enables the submission, publication, and retrieval of robust summaries of Quantitative Structure-Activity Relationship (QSAR) models. Standardised QSAR model summaries are provided by the QSAR Model Reporting Format (QMRF), which reports key information on the model and on any validation studies. The database has been developed and implemented by the European Commission’s Joint Research Centre Institute for Health & Consumer Protection IHCP. It represents a tailor-made web accessible system, based on building blocks from the AMBIT2 software for chemo informatics data management AMBIT.
AMBIT is an open source software for cheminformatics data management, developed by Nina Jeliazkova and Joanna Jaworska within the framework of a CEFIC LRI funded project. The AMBIT system consists of a database and functional modules allowing a variety of evaluations and data mining of the information, stored in the database. AMBIT’s database allows storage of chemical structures and their identifiers, experimental data and test descriptions, literature references, information about QSAR models and finally chemical structure attributes such as molecular descriptors. The database is based on a Relational Database Management System (RDBMS), which allows much faster and convenient access to the data in contrast to flat text files. Data stored in AMBIT can be searched in a number of ways: name, CAS, structure, Smiles, 2D fragment and structural similarity. A format for importing QMRF reports in the database, as well as a friendly user interface, facilitating data import, have been implemented.
QSAR Model Reporting Format (QMRF) is a template format developed at the Computational Toxicology and Modelling lab of the JRC's Institute for Health and Consumer Protection, located in Ispra, Italy, to facilitate information gathered for SAR and QSAR models. The JRC QSAR Model Database is a reference site for retrieving robust summaries of QSAR models.
In the regulatory assessment of chemicals (for example, under REACH), the QSAR models are playing an increasingly important role in predicting properties for hazard and risk assessment. This implies both a need to be able to identify relevant QSARs and to use them to derive estimates and/or have access to their pre-calculated estimates. It were these needs that prompted the development of a database of QSAR models (i.e. an inventory of information on the models) at JRC.
Description and availability
QMRF is a harmonised template for summarising and reporting key information on (Q)SAR models, including the results of any validation studies. The information is structured according to the OECD (Q)SAR validation principles. The JRC QSAR Model Database is freely accessible with the help of the following applications:
- QMRF Editor is a Java application, which provides user friendly interface for editing QMRF files.
- QMRF Inventory is a web application, designed to accommodate users to search QMRF files.
While QMRF Editor had used Java Web Start in the past, the current version is designed as a standalone desktop application. It is available for download from SourceForge. The JRC QSAR Model Database is accessible at Predictive Toxicology.
The JRC QSAR Model Database is implemented through servlets, Java Server Pages and JSP tags. Apache Tomcat or another servlet container is required to run the servlets and JSP. The core substructure search functionality is provided by the CDK chemoinformatics library. The relational database management system is MySQL. Custom JSP tags have been developed for similarity searching to facilitate the integration of AMBIT functionality within web pages.
Anonymous users can browse the QSAR Model Database for published QMRFs and perform document or substance searching. Registered users can save their searches. QMRF documents can be searched by applying a subset of the following criteria:
- Free text
- Free text (Boolean)
- Endpoint (predefined list)
- QMRF Number
Data stored in AMBIT can be searched in a number of ways: name, CAS, structure, Smiles, 2D fragment and structural similarity. The unique feature of AMBIT is the ability to store multifaceted information about chemical structures and provide a searchable interface linking these diverse components. Methods for assessing structural similarity, proximity in chemical space or commonality of mechanism of action can be used for QSAR applicability assessment and chemical grouping. The software relies on various open source software libraries and is an open source code to achieve maximum quality, transparency and ease of dissemination.
The following supportive documents can be found here:
- User Manual JRC QSAR Model Database
- Editorial Guidelines JRC QSAR Model Database
- Reviewer's protocol JRC QSAR Model Database
- Instructions how to create SDF files
Please do not hesitate to file reports for any software bug you find. The trouble ticketing system is available here.
To checkout the complete code tree with SVN use:
svn checkout svn://svn.code.sf.net/p/qmrf/code qmrf-code