Positive-unlabeled learning for Pv RV
Preface
1
Data engineering
1.1
purf
package installization
1.2
Retrieving
P. vivax
protein variables
1.2.1
Immunological data sets
1.2.2
Proteomic data sets
1.2.3
Structural data sets
1.2.4
Genomic data sets
1.2.5
Final assembly
1.3
Generating ML input
2
Model training
2.1
Hyper-parameter tuning for PURF
2.1.1
Analysis
2.1.1.1
Pv data set
2.1.1.2
Pf + Pv combined data set
2.1.2
Plotting
2.1.2.1
Pv data set
2.1.2.2
Bimodal distribution plot
2.1.2.3
Known antigen ranking
2.1.2.4
Pf + Pv combined data set
2.1.2.5
Bimodal distribution plot
2.1.2.6
Known antigen ranking
2.1.3
Results for positive level = 0.5
3
Model evaluation
3.1
Known antigen prediction accuracy
3.1.1
Analysis
3.1.1.1
Pf model cross-species predictions
3.1.1.2
Pv model cross-species predictions
3.2
Adversarial controls
3.2.1
Analysis
3.2.1.1
Pv data set
3.2.1.2
Pf + Pv combined data set
3.2.2
Plotting
3.2.2.1
Mean differences in scores
3.2.2.2
Investigation of known antigen groups
3.2.2.3
Labeled known antigen accuracies
4
Model comparisons
4.1
Known antigen score comparisons
4.2
Known antigen prediction summary
4.3
Score and species association
4.4
Unsupervised random forest analysis
4.4.1
Analysis
4.5
Tree depth analysis
4.5.1
Analysis
4.5.2
Plotting
5
Model interpretation
5.1
Proximity matrix calculation
5.1.1
Analysis
5.1.1.1
Pv data set
5.1.1.2
Pf + Pv combined data set
5.1.2
Plotting
5.1.2.1
Pf + Pv combined data set
5.2
Clustering of predicted positives
5.2.1
Analysis
5.3
Variable importance
5.3.1
Analysis
5.3.1.1
Pv data set
5.3.1.2
Pf + Pv combined data set
5.3.2
Plotting
5.3.2.1
Pv data set
5.3.2.2
Pf + Pv combined data set
5.3.3
Comparison of top 10 important variables
5.3.3.1
Venn diagram
5.3.3.2
Top 10 variables from combined model
6
Top candidate comparisons
6.1
Clustering analysis
6.1.1
Analysis
6.2
Gene ontology analysis
6.2.1
Analysis
6.2.2
Plotting
6.3
Candidate antigen characterization
6.3.1
Processing table
7
Summary of candidates
References
Published with bookdown
Malaria vaccine antigen identification for
Plasmodium vivax
References