Random Forest model for the genotoxicity classificaton of MWCNTs.
Random Forest model for the genotoxicity classificaton of MWCNTs.
Training dataset is available in this link
QMRF Report
1. QSAR Identifier
1.1.QSAR identifier (title):
Random Forest model for classifying genotoxicity of multi-walled carbon nanotubes (MWCNTs).
The model has been presented in the pubclication “Machine learning methods for multi-walled carbon
nanotubes (MWCNT) genotoxicity prediction", M. Kotzabasaki, I. Sotiropoulos, C. Charitidis, H. Sarimveis, Nanoscale Adv., 2021, 3(11), 3167-3176, DOI: 10.1039/d0na00600a
1.2.Other related models:
None
1.3.Software coding the model:
Jaqpot is a web platform that support development, validation and sharing of QSAR models apps.jaqpot.org
2. General Information
2.1.Date of QMRF:
28 September 2019
2.2.QMRF author(s) and contact details:
Iason Sotiropoulos, National Technical University of Athens, jasonsoti1@gmail.com
Marianna Kotzabasaki, National Technical University of Athens, mariannako@chemeng.ntua.gr
2.3.Date of QMRF update(s):
28 September 2019
2.4.QMRF update(s):
2.5.Model developer(s) and contact details:
[1] Iason Sotiropoulos, National Technical University of Athens, jasonsoti1@gmail.com
[2] Marianna Kotzabasaki, National Technical University of Athens, mariannako@chemeng.ntua.gr
2.6.Date of model development and/or publication:
The model was developed on 11/04/2020.
2.7.Reference(s) to main scientific papers and/or software package:
2.8.Availability of information about the model:
2.9.Availability of another QMRF for exactly the same model:
3. Defining the endpoint - OECD Principle 1
3.1.Species:
3.2.Endpoint:
Genotoxicity of MWCNTs.
3.3.Comment on endpoint:
The endpoint is classified in two classes, class 0 and 1, where each class corresponds into genotoxic and non-genotoxic MWCNTs respectivelly.
3.4.Endpoint units:
Dimensionless endpoint.
3.5.Dependent variable:
Genotoxicity of MWCNTs.
3.6.Experimental protocol:
3.7.Endpoint data quality and variability:
4. Defining the algorithm - OECD Principle 2
4.1.Type of model:
Logistic Regression (LR)
4.2.Explicit algorithm:
Logistic Regression (C=1374.54, penalty='l1',random_state=42)
4.3.Descriptors in the model:
[1] Purity - Percentage of carbon purity in the MWCNT (Continuous)
[2] Zeta average - Hydrodynamic size of MWCNTs measured in DLS at a dose of 12.5 μg/ml (Continuous)
[3]Length - The length of the MWCNTs(Continuous)
4.4.Descriptor selection:
4.5.Algorithm and descriptor generation:
4.6.Software name and version for descriptor generation:
4.7.Chemicals/Descriptors ratio:
5. Defining the algorithm - OECD Principle 3
5.1.Description of the applicability domain of the model:
5.2.Method used to assess the applicability domain: Leverage method
5.3.Software name and version for applicability domain assessment:
5.4.Limits of applicability:
Leverage Threshold value = 1.2
6. Internal Validation - OECD Principle 3
6.1.Availability of the training set:
Yes
6.2.Available information for the training set:
CAS RN: No
Chemical Name: Yes
Smiles: No
Formula: No
INChI: No
MOL file: No
NanoMaterial: No
6.3.Data for each descriptor variable for the training set:
All
6.4.Data for the dependent variable for the training set:
All
6.5.Other information about the training set:
6.6.Pre-processing of data before modelling:
The data are scaled in the range [0,1]
6.7.Statistics for goodness-of-fit:
Train Accuracy: 1.0
Train F1-Score: 1.0
Train MCC: 1.0
6.8.Robustness - Statistics obtained by 3-fold cross-validation:
6.9.Robustness - Statistics obtained by leave-many-out cross-validation:
Accuracy scores for 4-fold cross validation: [0.75, 0.75, 1.0, 0.67]
Cross Validation Mean Score: 0.792
6.10.Robustness - Statistics obtained by Y-scrambling:
6.11.Robustness - Statistics obtained by bootstrap:
6.12.Robustness - Statistics obtained by other methods:
7. External Validation - OECD Principle 4
7.1.Availability of the external validation set:
Yes
7.2.Available information for the external validation set:
CAS RN: No
Chemical Name: Yes
Smiles: No
Formula: No
INChI: No
MOL file: No
NanoMaterial: No
7.3.Data for each descriptor variable for the external validation set:
All
7.4.Data for the dependent variable for the external validation set:
All
7.5.Other information about the external validation set:
The extrernal validation test set consists of 33% (5 out of 15 samples) of the original data selected by the Kennard-Stone algorithm.
7.6.Experimental design of test set:
By setting aside 33% of the original pool of samples according to the Kennard-Stone algorithm.
7.7.Predictivity - Statistics obtained by external validation:
7.8.Predictivity - Assessment of the external validation set:
7.9.Comments on the external validation of the model:
8. Providing a mechanistic Interpetation - OECD Principle 5
8.1.Mechanistic basis of the model:
8.2.A priori or a posteriori mechanistic interpretation:
8.3.Other information about the mechanistic interpretation:
9. Miscellaneous information
9.1.Comments:
9.2.Bibliography:
9.3.Supporting information:
Training set(s)
Training dataset is available in this link
(https://app.jaqpot.org/dataset/wZD0GObM42fPMQAOXUwH7Z)
Test set(s)Supporting information
Model in PMML Format
<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
<PMML xmlns="http://www.dmg.org/PMML-4_4" xmlns:data="http://jpmml.org/jpmml-model/InlineTable" version="4.4">
<Header>
<Application name="JPMML-SkLearn" version="1.6.7"/>
<Timestamp>2021-01-13T12:52:54Z</Timestamp>
</Header>
<MiningBuildTask>
<Extension>PMMLPipeline(steps=[('scale', MinMaxScaler()),
('classifier', RandomForestClassifier(max_features=0.6664580622368363,
min_samples_split=0.11646759543664197, n_estimators=19,
random_state=42))])</Extension>
</MiningBuildTask>
<DataDictionary>
<DataField name="Genotoxicity" optype="categorical" dataType="integer">
<Value value="0"/>
<Value value="1"/>
</DataField>
<DataField name="Length ave. (nm) " optype="continuous" dataType="double"/>
<DataField name="Purity (%)" optype="continuous" dataType="double"/>
<DataField name="Zave (12.5 ug/ml)" optype="continuous" dataType="double"/>
</DataDictionary>
<TransformationDictionary/>
<MiningModel functionName="classification">
<MiningSchema>
<MiningField name="Genotoxicity" usageType="target"/>
<MiningField name="Length ave. (nm) "/>
<MiningField name="Purity (%)"/>
<MiningField name="Zave (12.5 ug/ml)"/>
</MiningSchema>
<Output>
<OutputField name="probability(0)" optype="continuous" dataType="double" feature="probability" value="0"/>
<OutputField name="probability(1)" optype="continuous" dataType="double" feature="probability" value="1"/>
</Output>
<LocalTransformations>
<DerivedField name="minMaxScaler(Length ave. (nm) )" optype="continuous" dataType="double">
<Apply function="+">
<Apply function="*">
<FieldRef field="Length ave. (nm) "/>
<Constant dataType="double">1.930092065391519E-4</Constant>
</Apply>
<Constant dataType="double">-0.10015247727316591</Constant>
</Apply>
</DerivedField>
<DerivedField name="minMaxScaler(Purity (%))" optype="continuous" dataType="double">
<Apply function="+">
<Apply function="*">
<FieldRef field="Purity (%)"/>
<Constant dataType="double">0.10869565217391301</Constant>
</Apply>
<Constant dataType="double">-9.78260869565217</Constant>
</Apply>
</DerivedField>
<DerivedField name="minMaxScaler(Zave (12.5 ug/ml))" optype="continuous" dataType="double">
<Apply function="+">
<Apply function="*">
<FieldRef field="Zave (12.5 ug/ml)"/>
<Constant dataType="double">0.001455604075691412</Constant>
</Apply>
<Constant dataType="double">-0.21106259097525473</Constant>
</Apply>
</DerivedField>
<DerivedField name="float(minMaxScaler(Length ave. (nm) ))" optype="continuous" dataType="float">
<FieldRef field="minMaxScaler(Length ave. (nm) )"/>
</DerivedField>
<DerivedField name="float(minMaxScaler(Purity (%)))" optype="continuous" dataType="float">
<FieldRef field="minMaxScaler(Purity (%))"/>
</DerivedField>
<DerivedField name="float(minMaxScaler(Zave (12.5 ug/ml)))" optype="continuous" dataType="float">
<FieldRef field="minMaxScaler(Zave (12.5 ug/ml))"/>
</DerivedField>
<DerivedField name="double(float(minMaxScaler(Length ave. (nm) )))" optype="continuous" dataType="double">
<FieldRef field="float(minMaxScaler(Length ave. (nm) ))"/>
</DerivedField>
<DerivedField name="double(float(minMaxScaler(Purity (%))))" optype="continuous" dataType="double">
<FieldRef field="float(minMaxScaler(Purity (%)))"/>
</DerivedField>
<DerivedField name="double(float(minMaxScaler(Zave (12.5 ug/ml))))" optype="continuous" dataType="double">
<FieldRef field="float(minMaxScaler(Zave (12.5 ug/ml)))"/>
</DerivedField>
</LocalTransformations>
<Segmentation multipleModelMethod="average">
<Segment id="1">
<True/>
<TreeModel functionName="classification" missingValueStrategy="nullPrediction">
<MiningSchema>
<MiningField name="double(float(minMaxScaler(Length ave. (nm) )))"/>
<MiningField name="double(float(minMaxScaler(Purity (%))))"/>
</MiningSchema>
<Node>
<True/>
<Node score="0" recordCount="7">
<SimplePredicate field="double(float(minMaxScaler(Length ave. (nm) )))" operator="lessOrEqual" value="0.05602092295885086"/>
<ScoreDistribution value="0" recordCount="7"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
<Node score="1" recordCount="5">
<SimplePredicate field="double(float(minMaxScaler(Purity (%))))" operator="lessOrEqual" value="0.6396739184856415"/>
<ScoreDistribution value="0" recordCount="0"/>
<ScoreDistribution value="1" recordCount="5"/>
</Node>
<Node score="0" recordCount="3">
<True/>
<ScoreDistribution value="0" recordCount="3"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
</Node>
</TreeModel>
</Segment>
<Segment id="2">
<True/>
<TreeModel functionName="classification" missingValueStrategy="nullPrediction">
<MiningSchema>
<MiningField name="double(float(minMaxScaler(Zave (12.5 ug/ml))))"/>
</MiningSchema>
<Node>
<True/>
<Node score="0" recordCount="11">
<SimplePredicate field="double(float(minMaxScaler(Zave (12.5 ug/ml))))" operator="lessOrEqual" value="0.1659388691186905"/>
<ScoreDistribution value="0" recordCount="11"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
<Node score="1" recordCount="4">
<True/>
<ScoreDistribution value="0" recordCount="0"/>
<ScoreDistribution value="1" recordCount="4"/>
</Node>
</Node>
</TreeModel>
</Segment>
<Segment id="3">
<True/>
<TreeModel functionName="classification" missingValueStrategy="nullPrediction">
<MiningSchema>
<MiningField name="double(float(minMaxScaler(Length ave. (nm) )))"/>
<MiningField name="double(float(minMaxScaler(Purity (%))))"/>
<MiningField name="double(float(minMaxScaler(Zave (12.5 ug/ml))))"/>
</MiningSchema>
<Node>
<True/>
<Node score="1" recordCount="4">
<SimplePredicate field="double(float(minMaxScaler(Purity (%))))" operator="lessOrEqual" value="0.8097826242446899"/>
<ScoreDistribution value="0" recordCount="0"/>
<ScoreDistribution value="1" recordCount="4"/>
</Node>
<Node>
<SimplePredicate field="double(float(minMaxScaler(Purity (%))))" operator="lessOrEqual" value="0.967391312122345"/>
<Node>
<SimplePredicate field="double(float(minMaxScaler(Purity (%))))" operator="lessOrEqual" value="0.9402174055576324"/>
<Node score="0" recordCount="1">
<SimplePredicate field="double(float(minMaxScaler(Length ave. (nm) )))" operator="lessOrEqual" value="0.07827488332986832"/>
<ScoreDistribution value="0" recordCount="1"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
<Node score="1" recordCount="1">
<True/>
<ScoreDistribution value="0" recordCount="0"/>
<ScoreDistribution value="1" recordCount="1"/>
</Node>
</Node>
<Node score="0" recordCount="6">
<True/>
<ScoreDistribution value="0" recordCount="6"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
</Node>
<Node score="0" recordCount="1">
<SimplePredicate field="double(float(minMaxScaler(Zave (12.5 ug/ml))))" operator="lessOrEqual" value="0.5072780204936862"/>
<ScoreDistribution value="0" recordCount="1"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
<Node score="1" recordCount="2">
<True/>
<ScoreDistribution value="0" recordCount="0"/>
<ScoreDistribution value="1" recordCount="2"/>
</Node>
</Node>
</TreeModel>
</Segment>
<Segment id="4">
<True/>
<TreeModel functionName="classification" missingValueStrategy="nullPrediction">
<MiningSchema>
<MiningField name="double(float(minMaxScaler(Purity (%))))"/>
<MiningField name="double(float(minMaxScaler(Zave (12.5 ug/ml))))"/>
</MiningSchema>
<Node>
<True/>
<Node score="0" recordCount="8">
<SimplePredicate field="double(float(minMaxScaler(Zave (12.5 ug/ml))))" operator="lessOrEqual" value="0.1659388691186905"/>
<ScoreDistribution value="0" recordCount="8"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
<Node score="1" recordCount="5">
<SimplePredicate field="double(float(minMaxScaler(Purity (%))))" operator="lessOrEqual" value="0.9402174055576324"/>
<ScoreDistribution value="0" recordCount="0"/>
<ScoreDistribution value="1" recordCount="5"/>
</Node>
<Node score="0" recordCount="2">
<True/>
<ScoreDistribution value="0" recordCount="2"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
</Node>
</TreeModel>
</Segment>
<Segment id="5">
<True/>
<TreeModel functionName="classification" missingValueStrategy="nullPrediction">
<MiningSchema>
<MiningField name="double(float(minMaxScaler(Purity (%))))"/>
<MiningField name="double(float(minMaxScaler(Zave (12.5 ug/ml))))"/>
</MiningSchema>
<Node>
<True/>
<Node>
<SimplePredicate field="double(float(minMaxScaler(Zave (12.5 ug/ml))))" operator="lessOrEqual" value="0.8296943306922913"/>
<Node score="1" recordCount="2">
<SimplePredicate field="double(float(minMaxScaler(Purity (%))))" operator="lessOrEqual" value="0.46195653080940247"/>
<ScoreDistribution value="0" recordCount="0"/>
<ScoreDistribution value="1" recordCount="2"/>
</Node>
<Node score="0" recordCount="11">
<True/>
<ScoreDistribution value="0" recordCount="11"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
</Node>
<Node score="1" recordCount="2">
<True/>
<ScoreDistribution value="0" recordCount="0"/>
<ScoreDistribution value="1" recordCount="2"/>
</Node>
</Node>
</TreeModel>
</Segment>
<Segment id="6">
<True/>
<TreeModel functionName="classification" missingValueStrategy="nullPrediction">
<MiningSchema>
<MiningField name="double(float(minMaxScaler(Purity (%))))"/>
</MiningSchema>
<Node>
<True/>
<Node score="1" recordCount="4">
<SimplePredicate field="double(float(minMaxScaler(Purity (%))))" operator="lessOrEqual" value="0.8043478429317474"/>
<ScoreDistribution value="0" recordCount="0"/>
<ScoreDistribution value="1" recordCount="4"/>
</Node>
<Node score="0" recordCount="11">
<True/>
<ScoreDistribution value="0" recordCount="11"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
</Node>
</TreeModel>
</Segment>
<Segment id="7">
<True/>
<TreeModel functionName="classification" missingValueStrategy="nullPrediction">
<MiningSchema>
<MiningField name="double(float(minMaxScaler(Purity (%))))"/>
</MiningSchema>
<Node>
<True/>
<Node score="1" recordCount="8">
<SimplePredicate field="double(float(minMaxScaler(Purity (%))))" operator="lessOrEqual" value="0.8043478429317474"/>
<ScoreDistribution value="0" recordCount="0"/>
<ScoreDistribution value="1" recordCount="8"/>
</Node>
<Node score="0" recordCount="7">
<True/>
<ScoreDistribution value="0" recordCount="7"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
</Node>
</TreeModel>
</Segment>
<Segment id="8">
<True/>
<TreeModel functionName="classification" missingValueStrategy="nullPrediction">
<MiningSchema>
<MiningField name="double(float(minMaxScaler(Length ave. (nm) )))"/>
<MiningField name="double(float(minMaxScaler(Purity (%))))"/>
<MiningField name="double(float(minMaxScaler(Zave (12.5 ug/ml))))"/>
</MiningSchema>
<Node>
<True/>
<Node score="0" recordCount="6">
<SimplePredicate field="double(float(minMaxScaler(Length ave. (nm) )))" operator="lessOrEqual" value="0.05602092295885086"/>
<ScoreDistribution value="0" recordCount="6"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
<Node>
<SimplePredicate field="double(float(minMaxScaler(Purity (%))))" operator="lessOrEqual" value="0.988586962223053"/>
<Node score="0" recordCount="1">
<SimplePredicate field="double(float(minMaxScaler(Zave (12.5 ug/ml))))" operator="lessOrEqual" value="0.0989810791797936"/>
<ScoreDistribution value="0" recordCount="1"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
<Node score="1" recordCount="6">
<True/>
<ScoreDistribution value="0" recordCount="0"/>
<ScoreDistribution value="1" recordCount="6"/>
</Node>
</Node>
<Node score="0" recordCount="2">
<True/>
<ScoreDistribution value="0" recordCount="2"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
</Node>
</TreeModel>
</Segment>
<Segment id="9">
<True/>
<TreeModel functionName="classification" missingValueStrategy="nullPrediction">
<MiningSchema>
<MiningField name="double(float(minMaxScaler(Purity (%))))"/>
<MiningField name="double(float(minMaxScaler(Zave (12.5 ug/ml))))"/>
</MiningSchema>
<Node>
<True/>
<Node score="0" recordCount="4">
<SimplePredicate field="double(float(minMaxScaler(Zave (12.5 ug/ml))))" operator="lessOrEqual" value="0.10116448532789946"/>
<ScoreDistribution value="0" recordCount="4"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
<Node score="1" recordCount="10">
<SimplePredicate field="double(float(minMaxScaler(Purity (%))))" operator="lessOrEqual" value="0.9402174055576324"/>
<ScoreDistribution value="0" recordCount="0"/>
<ScoreDistribution value="1" recordCount="10"/>
</Node>
<Node score="0" recordCount="1">
<True/>
<ScoreDistribution value="0" recordCount="1"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
</Node>
</TreeModel>
</Segment>
<Segment id="10">
<True/>
<TreeModel functionName="classification" missingValueStrategy="nullPrediction">
<MiningSchema>
<MiningField name="double(float(minMaxScaler(Length ave. (nm) )))"/>
<MiningField name="double(float(minMaxScaler(Purity (%))))"/>
</MiningSchema>
<Node>
<True/>
<Node score="0" recordCount="4">
<SimplePredicate field="double(float(minMaxScaler(Length ave. (nm) )))" operator="lessOrEqual" value="0.05214143730700016"/>
<ScoreDistribution value="0" recordCount="4"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
<Node score="1" recordCount="5">
<SimplePredicate field="double(float(minMaxScaler(Purity (%))))" operator="lessOrEqual" value="0.6396739184856415"/>
<ScoreDistribution value="0" recordCount="0"/>
<ScoreDistribution value="1" recordCount="5"/>
</Node>
<Node score="0" recordCount="4">
<SimplePredicate field="double(float(minMaxScaler(Length ave. (nm) )))" operator="lessOrEqual" value="0.6047171428799629"/>
<ScoreDistribution value="0" recordCount="4"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
<Node score="1" recordCount="2">
<True/>
<ScoreDistribution value="0" recordCount="0"/>
<ScoreDistribution value="1" recordCount="2"/>
</Node>
</Node>
</TreeModel>
</Segment>
<Segment id="11">
<True/>
<TreeModel functionName="classification" missingValueStrategy="nullPrediction">
<MiningSchema>
<MiningField name="double(float(minMaxScaler(Length ave. (nm) )))"/>
<MiningField name="double(float(minMaxScaler(Zave (12.5 ug/ml))))"/>
</MiningSchema>
<Node>
<True/>
<Node score="0" recordCount="8">
<SimplePredicate field="double(float(minMaxScaler(Length ave. (nm) )))" operator="lessOrEqual" value="0.10263264551758766"/>
<ScoreDistribution value="0" recordCount="8"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
<Node score="0" recordCount="1">
<SimplePredicate field="double(float(minMaxScaler(Zave (12.5 ug/ml))))" operator="lessOrEqual" value="0.2176128146238625"/>
<ScoreDistribution value="0" recordCount="1"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
<Node score="1" recordCount="5">
<SimplePredicate field="double(float(minMaxScaler(Length ave. (nm) )))" operator="lessOrEqual" value="0.4229024574160576"/>
<ScoreDistribution value="0" recordCount="0"/>
<ScoreDistribution value="1" recordCount="5"/>
</Node>
<Node score="0" recordCount="1">
<True/>
<ScoreDistribution value="0" recordCount="1"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
</Node>
</TreeModel>
</Segment>
<Segment id="12">
<True/>
<TreeModel functionName="classification" missingValueStrategy="nullPrediction">
<MiningSchema>
<MiningField name="double(float(minMaxScaler(Length ave. (nm) )))"/>
<MiningField name="double(float(minMaxScaler(Purity (%))))"/>
<MiningField name="double(float(minMaxScaler(Zave (12.5 ug/ml))))"/>
</MiningSchema>
<Node>
<True/>
<Node score="0" recordCount="8">
<SimplePredicate field="double(float(minMaxScaler(Zave (12.5 ug/ml))))" operator="lessOrEqual" value="0.1659388691186905"/>
<ScoreDistribution value="0" recordCount="8"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
<Node score="1" recordCount="5">
<SimplePredicate field="double(float(minMaxScaler(Purity (%))))" operator="lessOrEqual" value="0.9402174055576324"/>
<ScoreDistribution value="0" recordCount="0"/>
<ScoreDistribution value="1" recordCount="5"/>
</Node>
<Node score="0" recordCount="1">
<SimplePredicate field="double(float(minMaxScaler(Length ave. (nm) )))" operator="lessOrEqual" value="0.5191368628293276"/>
<ScoreDistribution value="0" recordCount="1"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
<Node score="1" recordCount="1">
<True/>
<ScoreDistribution value="0" recordCount="0"/>
<ScoreDistribution value="1" recordCount="1"/>
</Node>
</Node>
</TreeModel>
</Segment>
<Segment id="13">
<True/>
<TreeModel functionName="classification" missingValueStrategy="nullPrediction">
<MiningSchema>
<MiningField name="double(float(minMaxScaler(Length ave. (nm) )))"/>
<MiningField name="double(float(minMaxScaler(Purity (%))))"/>
<MiningField name="double(float(minMaxScaler(Zave (12.5 ug/ml))))"/>
</MiningSchema>
<Node>
<True/>
<Node score="0" recordCount="4">
<SimplePredicate field="double(float(minMaxScaler(Zave (12.5 ug/ml))))" operator="lessOrEqual" value="0.16011644899845123"/>
<ScoreDistribution value="0" recordCount="4"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
<Node score="0" recordCount="1">
<SimplePredicate field="double(float(minMaxScaler(Length ave. (nm) )))" operator="lessOrEqual" value="0.0508000236004591"/>
<ScoreDistribution value="0" recordCount="1"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
<Node score="1" recordCount="8">
<SimplePredicate field="double(float(minMaxScaler(Length ave. (nm) )))" operator="lessOrEqual" value="0.44036979228258133"/>
<ScoreDistribution value="0" recordCount="0"/>
<ScoreDistribution value="1" recordCount="8"/>
</Node>
<Node score="1" recordCount="1">
<SimplePredicate field="double(float(minMaxScaler(Purity (%))))" operator="lessOrEqual" value="0.988586962223053"/>
<ScoreDistribution value="0" recordCount="0"/>
<ScoreDistribution value="1" recordCount="1"/>
</Node>
<Node score="0" recordCount="1">
<True/>
<ScoreDistribution value="0" recordCount="1"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
</Node>
</TreeModel>
</Segment>
<Segment id="14">
<True/>
<TreeModel functionName="classification" missingValueStrategy="nullPrediction">
<MiningSchema>
<MiningField name="double(float(minMaxScaler(Length ave. (nm) )))"/>
<MiningField name="double(float(minMaxScaler(Zave (12.5 ug/ml))))"/>
</MiningSchema>
<Node>
<True/>
<Node score="0" recordCount="6">
<SimplePredicate field="double(float(minMaxScaler(Zave (12.5 ug/ml))))" operator="lessOrEqual" value="0.21106259524822235"/>
<ScoreDistribution value="0" recordCount="6"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
<Node score="0" recordCount="1">
<SimplePredicate field="double(float(minMaxScaler(Length ave. (nm) )))" operator="lessOrEqual" value="0.0974117461591959"/>
<ScoreDistribution value="0" recordCount="1"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
<Node score="1" recordCount="8">
<True/>
<ScoreDistribution value="0" recordCount="0"/>
<ScoreDistribution value="1" recordCount="8"/>
</Node>
</Node>
</TreeModel>
</Segment>
<Segment id="15">
<True/>
<TreeModel functionName="classification" missingValueStrategy="nullPrediction">
<MiningSchema>
<MiningField name="double(float(minMaxScaler(Length ave. (nm) )))"/>
<MiningField name="double(float(minMaxScaler(Purity (%))))"/>
</MiningSchema>
<Node>
<True/>
<Node score="1" recordCount="4">
<SimplePredicate field="double(float(minMaxScaler(Purity (%))))" operator="lessOrEqual" value="0.623369574546814"/>
<ScoreDistribution value="0" recordCount="0"/>
<ScoreDistribution value="1" recordCount="4"/>
</Node>
<Node score="0" recordCount="6">
<SimplePredicate field="double(float(minMaxScaler(Purity (%))))" operator="lessOrEqual" value="0.967391312122345"/>
<ScoreDistribution value="0" recordCount="6"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
<Node score="0" recordCount="4">
<SimplePredicate field="double(float(minMaxScaler(Length ave. (nm) )))" operator="lessOrEqual" value="0.8405743837356567"/>
<ScoreDistribution value="0" recordCount="4"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
<Node score="1" recordCount="1">
<True/>
<ScoreDistribution value="0" recordCount="0"/>
<ScoreDistribution value="1" recordCount="1"/>
</Node>
</Node>
</TreeModel>
</Segment>
<Segment id="16">
<True/>
<TreeModel functionName="classification" missingValueStrategy="nullPrediction">
<MiningSchema>
<MiningField name="double(float(minMaxScaler(Length ave. (nm) )))"/>
<MiningField name="double(float(minMaxScaler(Purity (%))))"/>
<MiningField name="double(float(minMaxScaler(Zave (12.5 ug/ml))))"/>
</MiningSchema>
<Node>
<True/>
<Node score="1" recordCount="2">
<SimplePredicate field="double(float(minMaxScaler(Purity (%))))" operator="lessOrEqual" value="0.8043478429317474"/>
<ScoreDistribution value="0" recordCount="0"/>
<ScoreDistribution value="1" recordCount="2"/>
</Node>
<Node>
<SimplePredicate field="double(float(minMaxScaler(Zave (12.5 ug/ml))))" operator="lessOrEqual" value="0.8296943306922913"/>
<Node score="0" recordCount="8">
<SimplePredicate field="double(float(minMaxScaler(Length ave. (nm) )))" operator="lessOrEqual" value="0.1251857727766037"/>
<ScoreDistribution value="0" recordCount="8"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
<Node score="0" recordCount="2">
<SimplePredicate field="double(float(minMaxScaler(Zave (12.5 ug/ml))))" operator="lessOrEqual" value="0.2176128146238625"/>
<ScoreDistribution value="0" recordCount="2"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
<Node score="1" recordCount="1">
<True/>
<ScoreDistribution value="0" recordCount="0"/>
<ScoreDistribution value="1" recordCount="1"/>
</Node>
</Node>
<Node score="1" recordCount="2">
<True/>
<ScoreDistribution value="0" recordCount="0"/>
<ScoreDistribution value="1" recordCount="2"/>
</Node>
</Node>
</TreeModel>
</Segment>
<Segment id="17">
<True/>
<TreeModel functionName="classification" missingValueStrategy="nullPrediction">
<MiningSchema>
<MiningField name="double(float(minMaxScaler(Length ave. (nm) )))"/>
<MiningField name="double(float(minMaxScaler(Zave (12.5 ug/ml))))"/>
</MiningSchema>
<Node>
<True/>
<Node score="0" recordCount="6">
<SimplePredicate field="double(float(minMaxScaler(Length ave. (nm) )))" operator="lessOrEqual" value="0.05602092295885086"/>
<ScoreDistribution value="0" recordCount="6"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
<Node score="0" recordCount="1">
<SimplePredicate field="double(float(minMaxScaler(Zave (12.5 ug/ml))))" operator="lessOrEqual" value="0.0989810791797936"/>
<ScoreDistribution value="0" recordCount="1"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
<Node score="1" recordCount="5">
<SimplePredicate field="double(float(minMaxScaler(Zave (12.5 ug/ml))))" operator="lessOrEqual" value="0.448326051235199"/>
<ScoreDistribution value="0" recordCount="0"/>
<ScoreDistribution value="1" recordCount="5"/>
</Node>
<Node score="0" recordCount="1">
<SimplePredicate field="double(float(minMaxScaler(Zave (12.5 ug/ml))))" operator="lessOrEqual" value="0.6310043632984161"/>
<ScoreDistribution value="0" recordCount="1"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
<Node score="1" recordCount="2">
<True/>
<ScoreDistribution value="0" recordCount="0"/>
<ScoreDistribution value="1" recordCount="2"/>
</Node>
</Node>
</TreeModel>
</Segment>
<Segment id="18">
<True/>
<TreeModel functionName="classification" missingValueStrategy="nullPrediction">
<MiningSchema>
<MiningField name="double(float(minMaxScaler(Length ave. (nm) )))"/>
<MiningField name="double(float(minMaxScaler(Purity (%))))"/>
</MiningSchema>
<Node>
<True/>
<Node score="0" recordCount="5">
<SimplePredicate field="double(float(minMaxScaler(Length ave. (nm) )))" operator="lessOrEqual" value="0.1251857727766037"/>
<ScoreDistribution value="0" recordCount="5"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
<Node score="1" recordCount="6">
<SimplePredicate field="double(float(minMaxScaler(Purity (%))))" operator="lessOrEqual" value="0.945652186870575"/>
<ScoreDistribution value="0" recordCount="0"/>
<ScoreDistribution value="1" recordCount="6"/>
</Node>
<Node score="0" recordCount="2">
<SimplePredicate field="double(float(minMaxScaler(Purity (%))))" operator="lessOrEqual" value="0.967391312122345"/>
<ScoreDistribution value="0" recordCount="2"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
<Node score="0" recordCount="1">
<SimplePredicate field="double(float(minMaxScaler(Length ave. (nm) )))" operator="lessOrEqual" value="0.8405743837356567"/>
<ScoreDistribution value="0" recordCount="1"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
<Node score="1" recordCount="1">
<True/>
<ScoreDistribution value="0" recordCount="0"/>
<ScoreDistribution value="1" recordCount="1"/>
</Node>
</Node>
</TreeModel>
</Segment>
<Segment id="19">
<True/>
<TreeModel functionName="classification" missingValueStrategy="nullPrediction">
<MiningSchema>
<MiningField name="double(float(minMaxScaler(Purity (%))))"/>
</MiningSchema>
<Node>
<True/>
<Node score="1" recordCount="8">
<SimplePredicate field="double(float(minMaxScaler(Purity (%))))" operator="lessOrEqual" value="0.9402174055576324"/>
<ScoreDistribution value="0" recordCount="0"/>
<ScoreDistribution value="1" recordCount="8"/>
</Node>
<Node score="0" recordCount="7">
<True/>
<ScoreDistribution value="0" recordCount="7"/>
<ScoreDistribution value="1" recordCount="0"/>
</Node>
</Node>
</TreeModel>
</Segment>
</Segmentation>
</MiningModel>
</PMML>