QMRF_v1.2_FishTox.xml

Extensible Markup Language (XML) icon QMRF_v1.2_FishTox.xml — Extensible Markup Language (XML), 21 KB (21955 bytes)

File contents

<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<!DOCTYPE QMRF PUBLIC "http://ambit.acad.bg/qmrf/qmrf.dtd" "qmrf.dtd">
<QMRF author="European Chemicals Bureau" contact="European Chemicals Bureau, IHCP, Joint Research Centre, European Commission" date="July 2007" email="ecb.qsar@jrc.it" name="(Q)SAR Model Reporting Format" schema_version="0.9" url="http://ecb.jrc.it/QSAR/" version="1.2">
<QMRF_chapters>
<QSAR_identifier chapter="1" help="" name="QSAR identifier">
<QSAR_title chapter="1.1" help="" name="QSAR identifier (title)">&lt;html&gt;&#13;
  &lt;head&gt;&#13;
&#13;
  &lt;/head&gt;&#13;
  &lt;body&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      QSAR model for narcosis (general) to fathead minnow, including non-polar &#13;
      (NPN) and polar (PN) narcosis.&#13;
    &lt;/p&gt;&#13;
  &lt;/body&gt;&#13;
&lt;/html&gt;&#13;
</QSAR_title>
<QSAR_models chapter="1.2" help="" name="Other related models"/>
<QSAR_software chapter="1.3" help="" name="Software coding the model">
			
      
<software_ref idref="firstsoftware"/>
</QSAR_software>
</QSAR_identifier>
<QSAR_General_information chapter="2" help="" name="General information">
<qmrf_date chapter="2.1" help="" name="Date of QMRF">&lt;html&gt;&#13;
  &lt;head&gt;&#13;
    &#13;
  &lt;/head&gt;&#13;
  &lt;body&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      21 May 2008&#13;
    &lt;/p&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      &#13;
    &lt;/p&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      &#13;
    &lt;/p&gt;&#13;
  &lt;/body&gt;&#13;
&lt;/html&gt;&#13;
</qmrf_date>
<qmrf_authors chapter="2.2" help="" name="QMRF author(s) and contact details">
		
      
<author_ref idref="firstauthor"/>
</qmrf_authors>
<qmrf_date_revision chapter="2.3" help="" name="Date of QMRF update(s)">&lt;html&gt;&#13;
  &lt;head&gt;&#13;
    &#13;
  &lt;/head&gt;&#13;
  &lt;body&gt;&#13;
  &lt;/body&gt;&#13;
&lt;/html&gt;&#13;
</qmrf_date_revision>
<qmrf_revision chapter="2.4" help="" name="QMRF update(s)">&lt;html&gt;&#13;
  &lt;head&gt;&#13;
&#13;
  &lt;/head&gt;&#13;
  &lt;body&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      &#13;
    &lt;/p&gt;&#13;
  &lt;/body&gt;&#13;
&lt;/html&gt;&#13;
</qmrf_revision>
<model_authors chapter="2.5" help="" name="Model developer(s) and contact details">
		
      
<author_ref idref="modelauthor"/>
</model_authors>
<model_date chapter="2.6" help="" name="Date of model development and/or publication">&lt;html&gt;&#13;
  &lt;head&gt;&#13;
&#13;
  &lt;/head&gt;&#13;
  &lt;body&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      The model was published for the first time in 2005 (Pavan et al., 2005), &#13;
      although the data set was known earlier.&#13;
    &lt;/p&gt;&#13;
  &lt;/body&gt;&#13;
&lt;/html&gt;&#13;
</model_date>
<references chapter="2.7" help="" name="Reference(s) to main scientific papers and/or software package">

      
<publication_ref idref="publications_catalog_2"/>
</references>
<info_availability chapter="2.8" help="" name="Availability of information about the model">&lt;html&gt;&#13;
  &lt;head&gt;&#13;
&#13;
  &lt;/head&gt;&#13;
  &lt;body&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      Non-proprietary.&#13;
    &lt;/p&gt;&#13;
  &lt;/body&gt;&#13;
&lt;/html&gt;&#13;
</info_availability>
<related_models chapter="2.9" help="" name="Availability of another QMRF for exactly the same model"/>
</QSAR_General_information>
<QSAR_Endpoint chapter="3" help="" name="Defining the endpoint - OECD Principle 1">
<model_species chapter="3.1" help="" name="Species">&lt;html&gt;&#13;
  &lt;head&gt;&#13;
&#13;
  &lt;/head&gt;&#13;
  &lt;body&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      Pimephales promelas (fathead minnow).&#13;
    &lt;/p&gt;&#13;
  &lt;/body&gt;&#13;
&lt;/html&gt;&#13;
</model_species>
<model_endpoint chapter="3.2" help="" name="Endpoint">

      
<endpoint_ref idref="endpoints_catalog_72"/>
</model_endpoint>
<endpoint_comments chapter="3.3" help="" name="Comment on endpoint">&lt;html&gt;&#13;
  &lt;head&gt;&#13;
&#13;
  &lt;/head&gt;&#13;
  &lt;body&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      LC50 - the concentration causing 50% lethality in Pimephales promelas, &#13;
      after an exposure of 96 hours.&#13;
    &lt;/p&gt;&#13;
  &lt;/body&gt;&#13;
&lt;/html&gt;&#13;
</endpoint_comments>
<endpoint_units chapter="3.4" help="" name="Endpoint units">&lt;html&gt;&#13;
  &lt;head&gt;&#13;
&#13;
  &lt;/head&gt;&#13;
  &lt;body&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      Moles per liter.&#13;
    &lt;/p&gt;&#13;
  &lt;/body&gt;&#13;
&lt;/html&gt;&#13;
</endpoint_units>
<endpoint_variable chapter="3.5" help="" name="Dependent variable">&lt;html&gt;&#13;
  &lt;head&gt;&#13;
&#13;
  &lt;/head&gt;&#13;
  &lt;body&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      log LC50&#13;
    &lt;/p&gt;&#13;
  &lt;/body&gt;&#13;
&lt;/html&gt;&#13;
</endpoint_variable>
<endpoint_protocol chapter="3.6" help="" name="Experimental protocol">&lt;html&gt;&#13;
  &lt;head&gt;&#13;
&#13;
  &lt;/head&gt;&#13;
  &lt;body&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      See: Veith et al. (1983) and Verhaar et al. (1995). &#13;
    &lt;/p&gt;&#13;
  &lt;/body&gt;&#13;
&lt;/html&gt;&#13;
</endpoint_protocol>
<endpoint_data_quality chapter="3.7" help="" name="Endpoint data quality and variability"/>
</QSAR_Endpoint>
<QSAR_Algorithm chapter="4" help="" name="Defining the algorithm - OECD Principle 2">
<algorithm_type chapter="4.1" help="" name="Type of model">&lt;html&gt;&#13;
  &lt;head&gt;&#13;
    &#13;
  &lt;/head&gt;&#13;
  &lt;body&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      Regression-based QSAR.&#13;
    &lt;/p&gt;&#13;
  &lt;/body&gt;&#13;
&lt;/html&gt;&#13;
</algorithm_type>
<algorithm_explicit chapter="4.2" help="" name="Explicit algorithm">
<algorithm_ref idref="algorithms_catalog_1"/>
<equation>&lt;html&gt;&#13;
  &lt;head&gt;&#13;
&#13;
  &lt;/head&gt;&#13;
  &lt;body&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      Log LC50 = -0.810 LogKow - 1.74&#13;
    &lt;/p&gt;&#13;
  &lt;/body&gt;&#13;
&lt;/html&gt;&#13;
</equation>
</algorithm_explicit>
<algorithms_descriptors chapter="4.3" help="" name="Descriptors in the model">
      
      
<descriptor_ref idref="descriptors_catalog_1"/>
</algorithms_descriptors>
<descriptors_selection chapter="4.4" help="" name="Descriptor selection">&lt;html&gt;&#13;
  &lt;head&gt;&#13;
&#13;
  &lt;/head&gt;&#13;
  &lt;body&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      &#13;
    &lt;/p&gt;&#13;
  &lt;/body&gt;&#13;
&lt;/html&gt;&#13;
</descriptors_selection>
<descriptors_generation chapter="4.5" help="" name="Algorithm and descriptor generation"/>
<descriptors_generation_software chapter="4.6" help="" name="Software name and version for descriptor generation" options="">
				
      
<software_ref idref="software_catalog_2"/>
</descriptors_generation_software>
<descriptors_chemicals_ratio chapter="4.7" help="" name="Descriptors/Chemicals ratio"/>
</QSAR_Algorithm>
<QSAR_Applicability_domain chapter="5" help="" name="Defining the applicability domain - OECD Principle 3">
<app_domain_description chapter="5.1" help="" name="Description of the applicability domain of the model">&lt;html&gt;&#13;
  &lt;head&gt;&#13;
&#13;
  &lt;/head&gt;&#13;
  &lt;body&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      The QSAR model was defined by the developer to be applicable to &#13;
      chemicals with logKow values in range from -1.31 to 6.20. HOWEVER, the &#13;
      model is not reccomended for chemicals with logKow &amp;lt; 1 due to large &#13;
      error for both non-polar and polar narcotics. It works best in the range &#13;
      3 &amp;lt; logKow &amp;lt; 6. The model can be suitable for chemicals acting by &#13;
      narcosis mode of action, and it is recomended only when the precise &#13;
      mechanism (i.e. non-polar or polar narcosis) is difficult to be &#13;
      determined. The structural domain includes aliphatic and aromatic &#13;
      hydrocarbons, halogenated aliphatic and aromatic hydrocarbons, ethers, &#13;
      alcohols, aromatic nitrocompounds, anilines and phenoles (only when a &#13;
      presence of an additional subtiuent on the benzene ring in case of &#13;
      phenols and anilines do not change the mechanism of action).&#13;
    &lt;/p&gt;&#13;
  &lt;/body&gt;&#13;
&lt;/html&gt;&#13;
</app_domain_description>
<app_domain_method chapter="5.2" help="" name="Method used to assess the applicability domain">&lt;html&gt;&#13;
  &lt;head&gt;&#13;
&#13;
  &lt;/head&gt;&#13;
  &lt;body&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      Expert judgement&#13;
    &lt;/p&gt;&#13;
  &lt;/body&gt;&#13;
&lt;/html&gt;&#13;
</app_domain_method>
<app_domain_software chapter="5.3" help="" name="Software name and version for applicability domain assessment">

      
<software_ref idref="software_catalog_3"/>
</app_domain_software>
<applicability_limits chapter="5.4" help="" name="Limits of applicability">&lt;html&gt;&#13;
  &lt;head&gt;&#13;
&#13;
  &lt;/head&gt;&#13;
  &lt;body&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      See 5.1&#13;
    &lt;/p&gt;&#13;
  &lt;/body&gt;&#13;
&lt;/html&gt;&#13;
</applicability_limits>
</QSAR_Applicability_domain>
<QSAR_Robustness chapter="6" help="" name="Internal validation - OECD Principle 4">
<training_set_availability answer="Yes" chapter="6.1" help="" name="Availability of the training set"/>
<training_set_data cas="Yes" chapter="6.2" chemname="Yes" formula="No" help="" inchi="No" mol="No" name="Available information for the training set" smiles="Yes"/>
<training_set_descriptors answer="All" chapter="6.3" help="" name="Data for each descriptor variable for the training set"/>
<dependent_var_availability answer="All" chapter="6.4" help="" name="Data for the dependent variable for the training set"/>
<other_info chapter="6.5" help="" name="Other information about the training set">&lt;html&gt;&#13;
  &lt;head&gt;&#13;
&#13;
  &lt;/head&gt;&#13;
  &lt;body&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      &#13;
    &lt;/p&gt;&#13;
  &lt;/body&gt;&#13;
&lt;/html&gt;&#13;
</other_info>
<preprocessing chapter="6.6" help="" name="Pre-processing of data before modelling"/>
<goodness_of_fit chapter="6.7" help="" name="Statistics for goodness-of-fit">&lt;html&gt;&#13;
  &lt;head&gt;&#13;
&#13;
  &lt;/head&gt;&#13;
  &lt;body&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      The model has been trained by 144 chemicals. &#13;
    &lt;/p&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      &#13;
    &lt;/p&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      Predictor=Constant  Coeff.= -1.74   SE=0.070;&#13;
    &lt;/p&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      Predictor=LogKow  Coeff.=-0.810   SE=0.026;&#13;
    &lt;/p&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      &#13;
    &lt;/p&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      The following fitness regression parameters were calculated for this &#13;
      QSAR:&#13;
    &lt;/p&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      R2 = 87.55 (Coefficient of determination);&#13;
    &lt;/p&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      R2adj = 87.46 (Coefficient of determination adjusted for the degrees of &#13;
      freedom);&#13;
    &lt;/p&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      s = 0.455 (Standard error of the estimate);&#13;
    &lt;/p&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      F =  998.3 (Fisher function);&#13;
    &lt;/p&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      LOF =  0.21 (Friedman modified);&#13;
    &lt;/p&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      SDEC =  0.452 (Standard Deviation Error in Calculation);&#13;
    &lt;/p&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      AIC =  0.213 (Akaike Information Criterion);&#13;
    &lt;/p&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      FIT =  6.83 (Kubinyi function).&#13;
    &lt;/p&gt;&#13;
  &lt;/body&gt;&#13;
&lt;/html&gt;&#13;
</goodness_of_fit>
<loo chapter="6.8" help="" name="Robustness - Statistics obtained by leave-one-out cross-validation">&lt;html&gt;&#13;
  &lt;head&gt;&#13;
&#13;
  &lt;/head&gt;&#13;
  &lt;body&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      The model was evaluated by leave-one-out internal cross-validation ( &#13;
      Q2LOO ).&#13;
    &lt;/p&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      Q2LOO = 87.06 (explained variance in prediction);&#13;
    &lt;/p&gt;&#13;
  &lt;/body&gt;&#13;
&lt;/html&gt;&#13;
</loo>
<lmo chapter="6.9" help="" name="Robustness - Statistics obtained by leave-many-out cross-validation"/>
<yscrambling chapter="6.10" help="" name="Robustness - Statistics obtained by Y-scrambling"/>
<bootstrap chapter="6.11" help="" name="Robustness - Statistics obtained by bootstrap">&lt;html&gt;&#13;
  &lt;head&gt;&#13;
&#13;
  &lt;/head&gt;&#13;
  &lt;body&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      The model was evaluated bootstrap with 5000 iterations.&#13;
    &lt;/p&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      Q2bootstrap = 87.11 (explained variance in prediction by bootstrapping); &#13;
    &lt;/p&gt;&#13;
  &lt;/body&gt;&#13;
&lt;/html&gt;&#13;
</bootstrap>
<other_statistics chapter="6.12" help="" name="Robustness - Statistics obtained by other methods">&lt;html&gt;&#13;
  &lt;head&gt;&#13;
&#13;
  &lt;/head&gt;&#13;
  &lt;body&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      SDEP = 0.461 (Standard Deviation Error in Prediction).&#13;
    &lt;/p&gt;&#13;
  &lt;/body&gt;&#13;
&lt;/html&gt;&#13;
</other_statistics>
</QSAR_Robustness>
<QSAR_Predictivity chapter="7" help="" name="External validation - OECD Principle 4">
<validation_set_availability answer="Yes" chapter="7.1" help="" name="Availability of the external validation set"/>
<validation_set_data cas="Yes" chapter="7.2" chemname="Yes" formula="No" help="" inchi="No" mol="No" name="Available information for the external validation set" smiles="Yes"/>
<validation_set_descriptors answer="All" chapter="7.3" help="" name="Data for each descriptor variable for the external validation set"/>
<validation_dependent_var_availability answer="All" chapter="7.4" help="" name="Data for the dependent variable for the external validation set"/>
<validation_other_info chapter="7.5" help="" name="Other information about the external validation set"/>
<experimental_design chapter="7.6" help="" name="Experimental design of test set">&lt;html&gt;&#13;
  &lt;head&gt;&#13;
&#13;
  &lt;/head&gt;&#13;
  &lt;body&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      For external validation, subset of a list containing 177 OECD SIDS &#13;
      chemicals (ENV/JM/TG(2004)26) with measured LC50 values to fathead &#13;
      minnow was used. The chemicals in the test set are within the range of &#13;
      log Kow of the training set and classified as narcotics (either &#13;
      non-polar or polar) in a consensus meachnism of action classification.&#13;
    &lt;/p&gt;&#13;
  &lt;/body&gt;&#13;
&lt;/html&gt;&#13;
</experimental_design>
<validation_predictivity chapter="7.7" help="" name="Predictivity - Statistics obtained by external validation">&lt;html&gt;&#13;
  &lt;head&gt;&#13;
&#13;
  &lt;/head&gt;&#13;
  &lt;body&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      N. ext = 17&#13;
    &lt;/p&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      Q2ext = 84.31&#13;
    &lt;/p&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      SDEPext = 0.637&#13;
    &lt;/p&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      The two nonyl phenols were excluded from these statistics due to high &#13;
      leverage.&#13;
    &lt;/p&gt;&#13;
  &lt;/body&gt;&#13;
&lt;/html&gt;&#13;
</validation_predictivity>
<validation_assessment chapter="7.8" help="" name="Predictivity - Assessment of the external validation set"/>
<validation_comments chapter="7.9" help="" name="Comments on the external validation of the model"/>
</QSAR_Predictivity>
<QSAR_Interpretation chapter="8" help="" name="Providing a mechanistic interpretation - OECD Principle 5">
<mechanistic_basis chapter="8.1" help="" name="Mechanistic basis of the model">&lt;html&gt;&#13;
  &lt;head&gt;&#13;
&#13;
  &lt;/head&gt;&#13;
  &lt;body&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      The mechanistic interpretation for logKow is related to the partitioning &#13;
      of the substance from water into the organisms lipid phase.&#13;
    &lt;/p&gt;&#13;
  &lt;/body&gt;&#13;
&lt;/html&gt;&#13;
</mechanistic_basis>
<mechanistic_basis_comments chapter="8.2" help="" name="A priori or a posteriori mechanistic interpretation">&lt;html&gt;&#13;
  &lt;head&gt;&#13;
&#13;
  &lt;/head&gt;&#13;
  &lt;body&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      Context dependent on the alert. &#13;
    &lt;/p&gt;&#13;
  &lt;/body&gt;&#13;
&lt;/html&gt;&#13;
</mechanistic_basis_comments>
<mechanistic_basis_info chapter="8.3" help="" name="Other information about the mechanistic interpretation"/>
</QSAR_Interpretation>
<QSAR_Miscelaneous chapter="9" help="" name="Miscellaneous information">
<comments chapter="9.1" help="" name="Comments">&lt;html&gt;&#13;
  &lt;head&gt;&#13;
&#13;
  &lt;/head&gt;&#13;
  &lt;body&gt;&#13;
    &lt;p style="margin-top: 0"&gt;&#13;
      Comment on the algorithm:   n = 144 (combination of the training sets of &#13;
      the models for non-polar narcosis and polar narcosis in EU TGD on risk &#13;
      assessment)&#13;
    &lt;/p&gt;&#13;
  &lt;/body&gt;&#13;
&lt;/html&gt;&#13;
</comments>
<bibliography chapter="9.2" help="" name="Bibliography">
				
      

<publication_ref idref="publications_catalog_1"/>
<publication_ref idref="publications_catalog_3"/>
</bibliography>
<attachments chapter="9.3" help="" name="Supporting information">
        
        
        
        
        
        
      


<attachment_training_data/>
<attachment_validation_data/>
<attachment_documents/>
</attachments>
</QSAR_Miscelaneous>
<QMRF_Summary chapter="10" help="" name="Summary (ECB Inventory)">
<QMRF_number chapter="10.1" help="" name="QMRF number">To be entered by ECB</QMRF_number>
<date_publication chapter="10.2" help="" name="Publication date">To be entered by ECB</date_publication>
<keywords chapter="10.3" help="" name="Keywords">To be entered by ECB</keywords>
<summary_comments chapter="10.4" help="" name="Comments">To be entered by ECB</summary_comments>
</QMRF_Summary>
</QMRF_chapters>
<Catalogs>
<software_catalog>
<software contact="" description="" id="firstsoftware" name="" number="" url=""/>
<software contact="" description="" id="software_catalog_2" name="" number="" url=""/>
<software contact="" description="" id="software_catalog_3" name="" number="" url=""/>
</software_catalog>
<algorithms_catalog>
<algorithm definition="Linear regression QSAR." description="QSAR model based on one single descriptor." id="algorithms_catalog_1" publication_ref=""/>
</algorithms_catalog>
<descriptors_catalog>
<descriptor description="The descriptor (log Kow) data are both experimental and calculated values. Even if Kow is usually considered a good physicochemical descriptor, there is no evidence that the measurements were made by the same protocol, in the same laboratory. Thus a certain amount of variability could be present." id="descriptors_catalog_1" name="Log Kow" publication_ref="" units=""/>
</descriptors_catalog>
<endpoints_catalog>
<endpoint group="3.Ecotoxic effects" id="endpoints_catalog_72" name="3.3.Acute toxicity to fish (lethality)" subgroup=""/>
</endpoints_catalog>
<publications_catalog>
<publication id="publications_catalog_1" title="Veith, GD, Call, DJ and Brooke, LT. (1983). Structure-toxicity relationships for the fathead minnow, Pimephales promelas: Narcotic industrial chemicals. Canadian Journal of Fisheries and Aquatic Sciences. 40, 743-748" url=""/>
<publication id="publications_catalog_2" title="Pavan M, Worth AP, Netzeva TI. 2005. Preliminary analysis of an aquatic toxicity dataset and assessment of QSAR models for narcosis. JRC Report No. 21749 EN.  " url="http://ecb.jrc.it/QSAR/"/>
<publication id="publications_catalog_3" title="Verhaar, H.J.M., Mulder, W. and Hermens, J.L.M. (1995). QSARs for ecotoxicity. In: Overview of structure-activity relationships for environmental endpoints, Part 1: General outline and procedure. Hermens, J.L.M. (Ed), Report prepared within the framework of the project “QSAR for Prediction of Fate and Effects of Chemicals in the Environment”, an international project of the Environmenta; Technologies RTD Programme (DGXII/D-1) of the European Commission under contract number EV5V-CT92-0211" url=""/>
</publications_catalog>
<authors_catalog>
<author affiliation="European Chemicals Bureau, TP 582, Institute for Health and Consumer Protection, Joint Research Centre, European Commission" contact="Via E.Fermi 2749, 21027 Ispra, Italy" email="manuela.pavan@ec.europa.eu" id="firstauthor" name="Manuela Pavan" number="" url="http://ecb.jrc.it/qsar/"/>
<author affiliation="European Chemicals Bureau, TP 582, Institute for Health and Consumer Protection, Joint Research Centre, European Commission" contact="Via E.Fermi 2749, 21027 Ispra, Italy" email="manuela.pavan@ec.europa.eu" id="modelauthor" name="Manuela Pavan" number="" url="http://ecb.jrc.it/qsar/"/>
</authors_catalog>
</Catalogs>
</QMRF>