Test Web Services
Descriptions and links to OpenTox web services
Please note that initial alpha-test services listed here are early development services and work in progress. They may be unavailable or not working! Also such test services may require technical QSAR or developer experience to use.
Alpha-Test Services
Entry Format:
Provider Name
Description Summary of Service Functionality:
Location:
Guidance:
Contact:
Date:
Version:
Bug Reporting:
PLEASE ADD YOUR NEW SERVICES TO THE INDIVIDUAL SERVICE TYPE PAGES (e.g., algorithm, model, integrated, etc.)
----------------------------------------------------------------------
Below is a list of the first iteration for current historical purposes only!
Services (Iteration 1.0)
Technical University of Munich
Description Summary of Service Functionality:
Location:
The web-browser readable entry page is located at: http://opentox.informatik.tu-muenchen.de:8080/OpenTox/Some of the available HTTP GET methods are also linked on that page.
Download:
The standalone version, tested on Linux and Windows XP, can be downloaded at
https://opentox.informatik.tu-muenchen.de/trac/TUMOpenTox/wiki/Download
Wiki and Bugtracker:
A documentation wiki and a bug tracker is available at
https://opentox.informatik.tu-muenchen.de/trac/TUMOpenTox/
Guidance:
A Linux example call (command line) to retrieve a list of available algorithm URIs would be:
curl -X GET -H 'Accept:text/uri-list' http://opentox.informatik.tu-muenchen.de:8080/OpenTox-dev/algorithm
Changing this to:
curl -X GET -H 'Accept:text/xml' http://opentox.informatik.tu-muenchen.de:8080/OpenTox-dev/algorithm
will give you the same information as above but formatted as XML.
You also can build (or train) models with the available classification and regression algorithms. Classification and regression algorithms at this stage of development only take WEKA arff files (URI to the file) as input, e.g.:
curl -X POST -d "dataset_id=http://someserver.com/files/iris.arff" http://opentox.informatik.tu-muenchen.de:8080/OpenTox-dev/algorithm/kNNclassification
Calculate descriptors: Momentarily the descriptor calculation services accept only SDF/MOL files as input for calculating descriptors. An example curl call would look like this:
curl -X POST -d "dataset_id=http://test.server.com/dataset/234/compound" -H 'Accept:text/xml' http://opentox.informatik.tu-muenchen.de:8080/OpenTox-dev/algorithm/CDKPhysChem
Here you can see that the HTTP media type that we expect in the response from the server is explicitly stated with text/xml.
An additional test server is available at http://opentox-dev.informatik.tu-muenchen.de:8080/OpenTox-dev. This server is used for internal testing and as a back-up in case the "production server" has problems.
Contact:
Tobias Girschick (TUM), Fabian Buchwald (TUM), Jörg Wicker (TUM)
in silico toxicology
Description Summary of Service Functionality:
Location:
Documentation, examples and source code on github account of Christoph Helma:
http://github.com/helma/opentox-compound
http://github.com/helma/opentox-feature
http://github.com/helma/opentox-dataset
http://github.com/helma/opentox-fminer
http://github.com/helma/opentox-lazar
The services run under the following base URIs (see github READMEs for
details):
http://webservices.in-silico.ch/compound/v0/
http://webservices.in-silico.ch/feature/v0/
http://webservices.in-silico.ch/dataset/v0/
http://webservices.in-silico.ch/fminer/v0/
http://webservices.in-silico.ch/lazar/v0/
Guidance:
At present, the servies provide YAML output and a few chemical structure
representations. XML output will come with the next update.
Please use the issue tracker on gihub to report problems and feature
requests.
Contact:
Christoph Helma (IST)
ALU-FR
Description Summary of Service Functionality:
Location:
http://opentox.informatik.uni-freiburg.de/validation
Guidance:
See
http://opentox.informatik.uni-freiburg.de/validation/examples
for curl examples.
Source code and some more documentation on github account of Martin Gütlein:
http://github.com/mguetlein/opentox-validation
Contact:
Martin Gütlein
Please use issue tracker to report bugs, request changes, etc:
http://github.com/mguetlein/opentox-validation/issues
Ideaconsult
Description Summary of Service Functionality:
OpenTox Compound, Feature, Dataset, Algorithm, Model , Ontology Web services.
Location:
http://apps.ideaconsult.net:8080/ambit2
http://apps.ideaconsult.net:8080/ontology
Guidance:
Contact:
National Technical Univeristy of Athens
Description Summary of Service Functionality:
1. Get XML representations of available algorithms
2. Build a classification model (SVM)
3. Build a regression model (SVM, MLR)
4. Validate a regression model (MLR)
5. Validate a classification model (SVM)
These service were built as a Java Web Application using the RESTlet 1.1.5
framework (http://www.restlet.org )
Location:
Home: http://opentox.ntua.gr:8080
Source Code: http://github.com/sopasakis/yaqp
Guidance:
You can find detailed instructions at http://opentox.ntua.gr:3003/guide
Contact:
Sopasakis Pantelis (NTUA) - chvng@mail.ntua.gr
Seascape Learning & Jawaharlal Nehru University
Description Summary of Service Functionality:
http://202.141.146.74:8080/MaxtoxMCSS
Guidance:
Contact:
Date:
Version:
Bug Reporting:
To get list of models available on MaxTox
|
In text form : curl -X GET -H "Accept:text" http://opentox2.informatik.uni-freiburg.de:8080/MaxtoxTest/model
In rdf/xml form: curl -X GET http://opentox2.informatik.uni-freiburg.de:8080/MaxtoxTest/model
|
To submit a dataset for prediction using model 1
|
curl -X POST -H "Content-Type:application/x-www-form-urlencoded" -d "dataset_uri=http://apps.ideaconsult.net:8080/ambit2/dataset/272?max=2" http://opentox2.informatik.uni-freiburg.de:8080/MaxtoxTest/model/1 |
Above command returns the prediction task Id, which can be used to retrieve the task status as follows
|
curl http://opentox2.informatik.uni-freiburg.de:8080/MaxtoxTest/task/Task_20110225121806
|
On completion of task, result dataset URI is returned which can then be seen on the dataset service server.
IBMC
Description Summary of Service Functionality:
IBMC - OpenTox Web services includes:
1) Caculation of MNA descritpors
2) Caculation of QNA descriptors
3) Build a regression model (SCR)
Location:
Home: http://195.178.207.160/OpenTox/Source code: http://ibmc.svn.sourceforge.net/viewvc/ibmc/OpenTox
Guidance:
The server is being developed on PHP&Delphi.
Example of curl commands to GET or POST RDF:
curl -X GET -H 'Accept:text/uri-list' http://195.178.207.160/OpenTox/algorithm
You also can build (or train) models with the SCR algorithms. SCR algorithm allows to get model on the basis of MNA or QNA descriptors.:
curl -X POST -d dataset_uri=https://ambit.uni-plovdiv.bg:8443/ambit2/dataset/2765 -d dataset_service=https://ambit.uni-plovdiv.bg:8443/ambit2/dataset -H subjectid:token http://195.178.207.160/OpenTox/Algorithm/MakeMNA
Creating model:
curl -iv -X POST -d dataset_uri=http://apps.ideaconsult.net:8080/ambit2/dataset/601261 -d prediction_feature=http://apps.ideaconsult.net:8080/ambit2/feature/534498 -H subjectid:token http://195.178.207.160/OpenTox/algorithm/MakeSCR
Predict values:
curl -X POST -d dataset_uri=http://apps.ideaconsult.net:8080/ambit2/dataset/601775 -d dataset_service=http://apps.ideaconsult.net:8080/ambit2/dataset http://195.178.207.160/opentox/model/makeSCR_15 -H subjectid:token
Contact:
Dmitry Druzhilovskiy dmitry.druzhilovsky@ibmc.msk.ru (IBMC)
Alexey Zakharov alexey.zakharov@ibmc.msk.ru (IBMC)Date:
2.02.2011
Version:
1.0
