OECD (Q)SAR Application Toolbox: Frequently Asked Questions

 

The set of Frequently Asked Questions on the Toolbox is available below.

 

 

List of Questions

 

 

 

 

 

Answers to the Questions

 

 

General


Q1-1: What does the Toolbox do?


The Toolbox is specifically designed to help the user to fill data gaps via the analogue approach or by building chemical categories according to the OECD Guidance on Grouping of Substances [OECD Series on Testing and Assessment No. 80. 2007. ENV/JM/MONO(2007)28]. This approach is used in a number of voluntary and regulatory programmes:

 

The Toolbox can help the user find adequate analogues to build a chemical category as well as to establish the arguments to build a category rationale (e.g. similarity of mechanisms of action, similarity of structural functional groups etc.). Specific guidance can be found in the Guidance Document for Using the OECD (Q)SAR Application Toolbox to Develop Chemical Categories According to the OECD Guidance on Grouping of Chemicals. It should be noted though that each programme mentioned above has its own reporting requirements and that the Toolbox does not overrule any of those reporting requirements. It should also be noted that the programmes mentioned above have requirements regarding the quality of the experimental data used for read-across and trend analysis and that the experimental results stored in the Toolbox will have to be verified by the user against those requirements (see also question “What quality assurance process is applied to the databases in the Toolbox?” ).

 

 

[Top of page ]

 

 

Q1-2: Will results obtained with the Toolbox be acceptable for regulatory purposes?

 

The acceptance of estimations obtained with (Q)SAR methodologies in regulatory settings is dependant of each individual regulatory system in the different member countries. The Report on the Regulatory Uses and Applications in OECD Member Countries of (Quantitative) Structure-Activity Relationships [(Q)SAR] Models in the Assessment of New and Existing Chemicals [OECD Series on Testing and Assessment No 58. 2006. ENV/JM/MONO(2006)25 ] provides some insight as to how (Q)SAR methodologies are used in member countries.

 

The Toolbox is specifically designed to help the user to fill data gaps via the analogue approach or by building chemical categories. This approach is used in a number of voluntary and regulatory programmes (see question “What does the Toolbox do?”). It should be noted that national regulatory systems have requirements regarding the quality of the experimental data used for read-across and trend analysis and that the experimental results stored in the Toolbox will have to be verified by the user against those requirements (see also question “What quality assurance process is applied to the databases in the Toolbox?” )

 

Q1-3: What endpoints does the Toolbox cover?

 

The Toolbox potentially covers all relevant regulatory endpoints. Nevertheless, the ability to fill data gaps via the analogue approach or by building chemical categories is limited by the availability of experimental results that can be used to perform read-across or trend analysis. Information on the availability of experimental data can be found below (see question: "What databases are included in the Toolbox?" ).


Q1-4: Does the Toolbox take into account aspects such as metabolism?

 

Yes. The Toolbox contains tools to estimate the occurrence of metabolites in different media. Guidance on how to use the metabolic profilers to e.g. refine a chemical category or to elaborate a metabolic pathway category are outlined in the Guidance Document for Using the OECD (Q)SAR Application Toolbox to Develop Chemical Categories According to the OECD Guidance on Grouping of Chemicals .

 

Q1-5: Can the Toolbox take into account mechanisms or modes of action of toxicity?

 

Yes. The Toolbox contains “mechanistic profilers” which can identify mechanisms or modes of actions relevant for different regulatory endpoints for a target chemical or a list of target chemicals. The user can then search for chemicals having the same mechanisms or modes of action to build chemical categories. Guidance on using these “mechanistic profilers” can be found either in the Guidance Document for Using the OECD (Q)SAR Application Toolbox to Develop Chemical Categories According to the OECD Guidance on Grouping of Chemicals, or among the published training material for the Toolbox.

 

 

[Top of page]

 

 

2. Databases


Q2-1: What databases are included in the Toolbox?

 

The Toolbox includes several databases as outlined in the table below.

 

 Data base

Content 

Developed and donated to the Toolbox by:

 Further information available in/at:

 

Canada

Bioaccumulation

Experimental results on bioaccumulation in aquatic organisms

Environment Canada

 


CEFIC - LRI BCF

Experimental results on bioaccumulation in fish

 CEFIC - LRI

 

Database

 

Danish EPA

 

 

Estimations for numerous properties and effects based on QSAR models

 

 Danish EPA

Danish EPA Database

ECETOC Aquatic Toxicity

 

Experimental results for aquatic toxicty

ECETOC

 

 

ECETOC 2003. Aquatic Hazard Assessment II, Technical Report No. 91. ISSN-0773-8072-91

 


ECETOC Eye Irritation

 

Experimental results on rabbit eye irritation 

 ECETOC

 

ECETOC 1992. Eye Irritation Reference Chemicals Data Bank, Technical Report No. 48. ISSN-0773-8072-48(2)

 


ECETOC Skin Sensitisation

 

Experimetnal results on skin and respirator sensitization

 ECETOC

 

ECETOC 1999. Skin and Respiratory Sensitisers: Reference Chemicals Data Bank. Technical report No. 77. ISSN-0773-8072-77

 

EPISUITE -
OBS - DATA

 


Experimental results on physical chemical aproperties, as accessed from EPISUITE. 

 

This is the PHYSROP database maintained at Syracuse Research Corporation


US - EPA & Syracuse Research Corporation

 

PHY5PROP Database 

 

ISSCAN update 3

 

Experimental results for genotoxicity and carcinogenicity

 

 

Instituto Superiore de Sanita, Italy

 

ISSCAN 1

 

ISSCAN 2

Japan Aquatic

 

 

Experimental results on aquatic toxicity based on tests performed within the Japanese Existing Chemicals Programme

 

 

Japanese Ministry of Environment

 

 Database

Japan EXCHEM

 

 

Experimental results from single dose toxicity test and mutagenicity test results performed under Japan's Exisiting Chemicals Programme

 

Japanese Ministry of Health, Labour and Welfare

 

 Database

OASIS Aquatic

 

Experimental results for aquatic toxicity

 

 

Laboratory of Mathematical Chemistry, Bulgaria

US-EPA

University of Knoxville, Tenessee

METI Japan

 

 

OASIS

Bioaccumulation

 

 

Experimental results on bioaccumulation

 

 

Laboratory of Mathematical Chemistry, Bulgaria

 

METI Japan

 

Exxon Mobile

 Database

OASIS

Biodegradation

 

Experimental results biodegradation

 

 

Laboratory of Mathematical Chemistry, Bulgaria

 

METI Japan

 

Database

OASIS ERBA

 

 

Estrogen Receptor Binding Affinity, expressed as relative binding affinities in comparison with the estradiol affinity

 

Laboratory of Mathematical Chemistry, Bulgaria

 

 

OASIS Genetox

 

Experimental results for genotoxicity

 

 

Laboratory of Mathematical Chemistry, Bulgaria

 

 

OASIS Skin Sensitisation

 

Experimental results for skin sensitisation

 

 

Laboratory of Mathematical Chemistry, Bulgaria

 

Unilever

 

Exxon Mobile

 

 

RIVM Skin Irritation

 

Primary Skin Irritation Indices from skin irritation tests

 

RIVM, the Netherlands

 

 

The database contains results from the following sources:

• Barratt, ECVAM Workshop 6: Corrosivity

 

• ECETOC 1995.Skin Irritation and Corrosion: Reference Chemicals Data Bank. Technical Report No.66

 

• Kodithala et al. (2002) Toxicol.Sci. 66(336-346)

US - EPA ECOTOX

 

 

Experimental results for aquatic and terrestrial toxicity as well as bioconcentration. This database integrates three previously independent databases - AQUIRE, PHYTOTOX, and TERRETOX

 

US - EPA

 

Database

 

 

It should be noted that these databases are considered to be based around reliable data but they have not been formally quality assured by OECD [see also question Q2-4 ].

 

 

[Top of page]

 

 

Q2-2: Can the Toolbox be extended with a user’s own data?


Any user can add his/her own data into his/her Toolbox installation. Indeed a user-friendly import function for experimental results is available within the Toolbox. Additional databases can therefore be imported into individual Toolbox installations by anybody and shared with other users of the Toolbox, independent of the OECD (Q)SAR project.  Specific guidance on importing databases is available.

 

A user can also import measured parameters (independent variables) to perform linear interpolation for an endpoint of interest.  Specific guidance is provided as part of the guidance document on "Tips and tricks."


 

Q2-3: Can I submit a database to OECD to be incorporated into the Toolbox as part of its core databases?


Yes. Proposals for databases to be included into the Toolbox as part of its core databases are handled by a formal request, which can be made by any stakeholder. The decision to include a database into the Toolbox is made by the QSAR Application Toolbox Management Group after considering the principles below:

• The need for the database. For example, there are currently more aquatic toxicity databases than terrestrial toxicity or health effects databases.  Therefore, there is a greater need for the latter two types of databases.  Following a similar line of reasoning there will be more value in expanding existing databases or adding similar databases (e.g., another fish species) which expand the biological or chemical domain, rather than adding a database that focuses on an already well represented organism or chemical class. Furthermore, the type of data in the additional databases should be restricted to the endpoints used for regulatory purposes or endpoints that can be directly linked to a regulatory endpoint by a toxic pathway.


• The quality of the documentation of the data in the database. Databases with information on test conditions and original references will be preferred. The aim is to provide sufficient information to allow the user to come to a conclusion regarding the quality of the data.


• The cost of adding and maintaining the database. Databases that are in a format ready for easy import into the Toolbox require minimal time and expense to incorporate. Other datasets may need to be put in an electronic format that would imply resources.  Depending on the funds available, the costs could be borne by the OECD.  Alternatively a member country or other stakeholder could offer to sponsor the effort and bear the cost as a contribution to the Toolbox.  In any of these cases, long-term maintenance and updating solutions will have to be considered.

The proposal should contain a description of the review that the data has undergone (e.g. peer-review by governmental authorities; other peer-review; no peer-review).  The proposal will also need to address maintenance and update of the database and contain a contact point responsible for providing the Secretariat with updates of the databases as appropriate.  All databases donated by a third party shall be given to the public domain, but any intellectual property associated with it will remain with the donor, and the OECD shall not infringe on the rights of the donors to use or dispose of these databases as they see fit.

 

Q2-4: What quality assurance process is applied to the databases in the Toolbox?


The resident databases of the Toolbox have been incorporated into the Toolbox as they have been donated. The OECD does not formally provide any quality assurance of data within the Toolbox. Therefore, whilst OECD expects these databases to include reliable data it does not take any responsibility for their use.
Nevertheless, as far as available, the Toolbox provides information on the quality assurance process performed by the donator, as well as full references of experimental results published in the open literature. This should allow the user to decide whether the data is adequate for their purposes.

 

 

[Top of page]

 

 

3. Profilers


Q3-1: Can I build my own profilers?


Yes. The Toolbox has a user friendly editing function, allowing the user to build his/her own profilers. Guidance on how to proceed is available in the form of a slide show as part of the training material published for the Toolbox

 

Q3-2: Can I submit my own profilers to OECD to be incorporated into the Toolbox?


The main purpose of the profilers of the Toolbox is to identify structural and mechanistic properties of a chemical or a list of chemicals, so that chemicals can be grouped into toxicological meaningful categories. During phase 2 of the development of the Toolbox, additional profilers will be developed, so that eventually the categories approach works uniformly for all discrete chemicals and for all regulatory endpoints. This activity will be steered by the OECD QSAR Application Toolbox Management Group. Proposals for inclusions of other profilers into the Toolbox can be handled by a formal request by any stakeholder.  The decision to include a profiling tool into the Toolbox is made by the OECD QSAR Application Toolbox Management Group after considering the principles below:

• The need for the profiling tool. The profiling tools or categorisation mechanisms which are currently implemented in the toolbox are most relevant for a limited number of endpoints (e.g. aquatic toxicity, genotoxicity, sensitisation, irritation). Any categorisation methods relevant for additional endpoints, and in particular systemic toxicity endpoints will have highest priority for inclusion into the Toolbox.


• The quality of the documentation for the method. Categorisation methods in the Toolbox should be fully documented, so as to provide full scientific background information to the user.

 

 

[Top of page]

 

 

4. QSAR models


Q4-1: Can I submit my own QSAR model to OECD to be incorporated into the Toolbox as part of its core library?

The Toolbox facilitates building chemical categories and using read-across and trend analysis to fill data gaps. In addition, the Toolbox contains a library of external (Q)SAR models which can also be used to fill data gaps. The OECD is not actively seeking external (Q)SAR models to add into the Toolbox, but will rely on proposals by member countries and other stakeholders. Proposals for inclusions of (Q)SAR models into the Toolbox are handled by a formal request by any stakeholder.  The decision to include a (Q)SAR model into the Toolbox is made by the OECD QSAR Application Toolbox Management Group after considering the principles below:


• The need for the model. For example, there are currently more models for estimating aquatic toxicity than terrestrial toxicity or health effects.  Therefore, there is a greater need for the latter two types of models.  Following a similar line of reasoning there will be more value in adding models which expand the applicability domain of existing models, rather than adding a model that focuses on an already well represented endpoint or chemical class. Furthermore, models for physical-chemical properties (e.g. pKa) which are useful for subcategorizing chemicals would have a high priority. In general the type of (Q)SAR models should be restricted to the endpoints used for regulatory purposes or endpoints that can be directly linked to a regulatory endpoint by a toxic pathway. 


• The quality of the documentation for the model. (Q)SAR models in the Toolbox should be documented according to the Guidance Document on the Validation of (Q)SAR Models (see Organisation for Economic Cooperation and Development. 2007. Guidance Document on the Validation of (Quantitative) Structure –Activity Relationships. OECD Environment Health and Safety Publications. Series on Testing and Assessment. No 69. ENV/JM/MONO(2007)2 ). The European Commission has developed a software that can be used to document the validation of (Q)SAR models in line with the OECD guidancedocument.


• The type of model. Experience with the Toolbox has shown that local mechanism-driven models have a higher regulatory acceptance than global statistical models. The former type of models would therefore have a higher priority for inclusion into the Toolbox.


• The cost of adding and maintaining the model. Models based on existing Toolbox descriptors will require minimal time and expense to incorporate into the Toolbox. In contrast, models requiring descriptors currently outside the Toolbox will require additional time and resources to incorporate into the Toolbox. Depending on the funds available, the costs could be borne by the OECD. Alternatively a member country or other stakeholder represented at the Joint Meeting could offer to sponsor the effort and bear the cost as a contribution to the Toolbox.  In any of these cases, long-term maintenance and updating solutions will have to be considered.


The European Commission is currently developing an inventory of (Q)SAR models   with detailed documentation according to the OECD Guidance Document on the Validation of (Q)SAR Models.  This inventory will be a source of models to be considered for inclusion into the Toolbox.


It should be noted that models using descriptors generated with proprietary software cannot be incorporated into the Toolbox, as the Toolbox is distributed freely. All models donated by a third party shall be given to the public domain, but any intellectual property associated with it will remain with the donor, and the OECD shall not infringe on the rights of the donors to use or dispose of these models as they see fit. 

 

 

[Top of page]

 

 

5. Installation


Q5-1: What do I need to install the Toolbox?


The Toolbox requires Microsoft Windows 2000, XP or Vista. It needs a PC with a processor running at 2GHz or faster, 3GB of hard disc space for storage and 1 GB (2GB for Vista) of RAM to operate. Administrative rights are also required during installation. However, once installed, normal user rights will be sufficient for running the application.


Q5-2: What is a USER folder? What is it used for?


The USER folder is used across the Toolbox for:
• Storing newly built QSAR models
• Default folder when importing databases into Toolbox
• Storing of config files when creating (importing) a new database
• Storing of new profiling schemes
• Temporary storage for the history report


These features rely on a USER folder being present so the Toobox would not function without a USER folder.

 

 

[Top of page]

 


Q5-3: What can I do if the InterBase service has insufficient rights?


The Toolbox has a number of databases that come with the installation package that use the InterBase RDBMS. In order for them to work InterBase server needs to be operational and have access to the database files.

It is possible that the InterBase service runs in a security context that does not allow it access to the database otherwise available to the currently logged in user. In this case you may get one of the following errors:


• Cannot establish connection to [FILENAME].
• Win32 error 5.
• Access is denied.


One way of solving that is to assure that the security context, that the InterBase service runs on, has the necessary access permissions for reading and writing to the database file.

Another work-around is to start the InterBase server in the same security context as your user. Assuming that you have full control over the DB files, so would the InterBase Server. To do that you need to:

1) Start the InterBase Server Manager located under Start Menu->Programs->InterBase 6 Open Edition

 

 

 

2) Uncheck the "Run InterBase server as a service on Windows NT" checkbox. 

 

  

 

 

[Top of page]

 

 

Q5-4: Can I access an InterBase database from a network drive?


The InterBase server can only access databases on the local machine. If you try to open a database located on a network drive you would receive the following message:

                 "Cannot establish connection to [FILENAME] cannot attach to password database"

 

There is no fix for this issue in the current version of the Toolbox, so in order to open the database you would need to move the ODB file to one of your local drives and edit the corresponding Config file's DBName row to point to the new database file's position.

 

Q5-5: How do I have to format a text file so that it can be used to store user lists of chemicals for the Toolbox?


 

Toolbox can read '*.smi' files (tab delimited text files) in two formats:
• Containing only SMILES in each row Example:
  CCC
  CCCC
  CCCCC


• Containing CAS#<TAB>Chemical_Name<TAB>SMILES Example (second name is missing):
  12345  Name1    CCC
  12                       CCCC
  234      Name3    CCCCC

 

Q5-6: I get a 'Failed to get procedure address' error message. How do I fix this?

 

The Toolbox needs a Java Virtual Machine to run some of its profilers. This message presents itself when the Java Virtual Machine (JVM) installed on the computer is an old version. Please install the version distributed with the Toolbox package - 'Java(TM) 6 Update 10' from Sun Microsystems, Inc.

 

[Top of page]


 

 

 

 

 

 

 

 

Top of page

Tool


Database