ETOX ProjectETOX ProjectETOX ProjectETOX ProjectETOX ProjectETOX Project
  Home > Results
ETOX Project Print page ETOX Project Send page

 

Human Outcomes Module


ETOX Project line


LIBRARY


ETOX Project line


ONTOBROWSER


ETOX Project line


eTOXlab


ETOX Project line


LiMTox


ETOX Project line


Collector


ETOX Project line


Standardiser


ETOX Project line


Verification


ETOX Project line


 

 

 

 

 

 

 

 

ETOX Project

 

 

 

 

ETOX Project

 

 

 

 

 

 

 

 

ETOX Project

 

 

 

 

 

 

 

 

 

ETOX Project

 

 

ETOX Project

 

 

 

 

ETOX Project

 

 

 

ETOX Project

 

 

 

ETOX Project

eTOX TOOLS

 

The eTOX tools freely accesible are the following:

 

Human Outcomes Module

 

Human Outcomes Module is a webservice developed by ERASMUS, in collaboration with FIMIM, that allows the search of similar marketed drugs from a simple SMILES code and retrieves adverse events data associated to these drugs from three sources currently implemented: literature (PubMed), EURETOS and LAERTES. The eTOXsys architecture was adapted to activate the eTOX Human Outcomes application directly from the eTOXsys GUI. The SMILES selected in eTOXsys is automatically passed on to the eTOX Human Outcomes prototype by means of a button that is shown in the new eTOXsys front-end

 

Please visit the Human Outcomes Module here

 

 

LIBRARY

 

From the beginning of the project, the eTOX consortium has been screening the literature and the World Wide Web looking for relevant data related to drug toxicity. This section aims to supply easy access to a chronological selection of recommended Articles, a list of selected Journals extracted from ISI Web of Knowledge with Toxicology filtering, and finally, a section of interesting Links, where all these resources are presented here by alphabetic order and organized by categories. The information here is updated in a monthly basis. An article describing the contents of this resource and the methodology used to compile them has been published in the Special Issue on Computational Toxicology of the Molecular Informatics journal. Please cite this resource as follows: ‘The eTOX library of public resources for in silico toxicity prediction’ M. Cases, M. Pastor and F.Sanz. Mol Inf 2013;32:24-35.

 

Please visit the eTOX Library here

 

 

ONTOBROWSER

 

As part of Novartis’ ongoing open source commitment (http://opensource.nibr.com), the company has recently published OntoBrowser as an open source software project. The source code has been released under the Apache License Version 2.0 and is available on GitHub (http://github.com/Novartis/ontobrowser). Consortium members are free to download and deploy the tool within their own environments. The release includes an installation guide to facilitate installation. As an open source project, consortium members are also free to modify the software to meet their own specific requirements.

The OntoBrowser tool was developed to manage ontologies and controlled terminologies (e.g. CDISC SEND) used within the scope of the eTOX project. The primary objective of the tool is to provide an online collaborative solution for expert curators to map report terms (from the eTOX database) to preferred ontology (or controlled terminology) terms. Additional features include: visualisation of ontologies in hierarchical/graph format, advanced search capabilities, peer review/approval workflow and web service access to data.

 

Please visit the eTOX OntoBrowser here

 

 

eTOXlab

 

eTOXlab is a flexible modeling framework. It was developed for supporting models predicting the biological properties of chemical compounds (e.g. QSAR models) in production environments. It is distributed embedded within a virtual machine containing all required software for building models and using them for prediction. Together, the ensemble eTOXlab-VM constitutes a portable, self-contained prediction engine that can be easily distributed and installed.

 

Please visit the eTOXlab landing page here

 

 

LiMTox

 

The LiMTox system is a text mining approach that extracts associations between compounds and toxicological end points at various levels of granularity and evidence types, all inspired by the content of toxicology reports. It integrates direct retrieval of associations between compounds and hepatotoxicity through combination of heterogeneous complementary strategies from term co-occurrences, linguistic rules, and text patterns to machine learning based text classification. It also returns relations of compounds and CYPs and compounds and biochemical markers (e.g. liver enzymes).

 

Please visit the LiMTox here

 

 

Collector

 

Collector is a tool that allows extracting from the Open PHACTS Discovery platform and the eTOX project series of compounds annotated with experimental data that can be used directly for building QSAR predictive models. Collector implements a flexible collection of filters for selecting the target, the biological data ranges and the physicochemical properties of the compounds.

Collector was intended to be installed locally and implements a local database for keeping record of the data extracted. The application provides a web based user a command line user interface.

Collector is being developed at the Pharmacoinformatics group (PhI), of the Research Programme on Biomedical Informatics (GRIB), IMIM/UPF, located in Barcelona

 

Please visit the Collector here

 

 

Standardiser

 

The Standardiser is a tool designed to provide a simple way of standardising molecules as a prelude to e.g. molecular modelling exercises. The tool is available on GitHub. The complete standardisation procedure is available as a Python module along with separate modules implementing each of the individual steps to allow generation of custom standardisation pipelines. The code used by this tool is released under the Apache 2.0 license created by EMBL-EBI.

 

Please visit the Standardiser here

 

 

Verification

 

The verification of the models developed within the eTOX project comprises an assessment of (i) the quality of data used to build (and test) the model, (ii) whether the predictions generated by the model when executed in eTOXsys are consistent, and (iii) the completeness of the documentation accompanying the model. This process has been semi-automated by implementing the required checks into a KNIME workflow and results in a more efficient and standardised verification process. The workflow requires the training data comprising structure ID, SMILES, experimental activity and predicted activity, along with specified outputs from eTOXsys. The result of the workflow is a PDF report that includes the verification status (Verified or Not Verified) and details on all of the verifications checks which were performed.

 

Please visit the Verification here