-
Notifications
You must be signed in to change notification settings - Fork 8
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Merge pull request #1 from Nardus/dev
Merge Dev
- Loading branch information
Showing
37 changed files
with
10,760 additions
and
102 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,64 @@ | ||
| # Numerical values represented in manuscript figures | ||
|
|
||
| ## Files included | ||
|
|
||
| | Figure | Panel | File name | Description | | ||
| |---------|-------|-----------------------------|-------------------------------------------------------------------------------------------------------------------------| | ||
| | Fig 1 | A | `fig1_a.csv` | AUC values for replicate train/calibrate/test splits | | ||
| | | B | `fig1_b_replicates.csv.zip` | Predictions from 1000 replicate train/calibrate/test splits, used to generate AUC curves | | ||
| | | | `fig1_b_bagged.csv` | Predictions from the bagged model, used to generate the summary AUC curve | | ||
| | | C | `fig1_c.csv` | Confusion matrix for the best model, derived from S1 Table / bagged predictions | | ||
| | | D | `fig1_d.csv` | Viruses encountered and cumulative success when screening viruses in the order recommended by the best model | | ||
| | Fig 2 | A | `fig2_a.csv` | SHAP values for each virus-genome feature combination, used to cluster viruses by explanation simililarity | | ||
| | | B | `fig2_b.csv` | Mean absolute SHAP values across all viruses, as a measure of feature importance | | ||
| | | C | `fig2_c.csv` | Feature importance and relative ranks for features measuring the same quantity in both referenced and unreferenced form | | ||
| | | D | `fig2_d_clusters.csv` | Relative importance of different clusters of correlated features (bar heights and error-bars in Fig 2D) | | ||
| | | | `fig2_d_featuresets.csv` | Composition of clusters (colours in Fig 2D) | | ||
| | Fig 3 | | -- | See data for S11 Fig; more detailed host/vector information recorded in [InternalData/NovelVirus_Hosts_Curated.csv](../InternalData/NovelVirus_Hosts_Curated.csv) | | ||
| | Fig 4 | | -- | These plots show partial residuals from a model fit - see the plotting script linked below | | ||
| | Fig 5 | A | -- | Values included in S1 Table | | ||
| | | B | `fig5_b.csv` | Predictions for different *Sarbecovirus* genomes | | ||
| | S1 Fig | | `s1_fig.csv` | List of viruses used for model training, including taxonomy | | ||
| | S2 Fig | A | `s2_fig_a.csv` | Predictions from the taxonomy-based model | | ||
| | | B | `s2_fig_b.csv` | Predictions from the phylogenetic neighbourhood-based model | | ||
| | S3 Fig | | `s3_fig.csv` | Family-specific AUC values, measuring the probability of accurately ranking known human-infecting viruses above other viruses from the same family | | ||
| | S4 Fig | | -- | Values included in S1 Table | | ||
| | S5 Fig | | -- | Values included in S1 Table | | ||
| | S6 Fig | | `s6_fig_taxonomy.csv` | Cumulative success when screening viruses in the order recommended by the taxonomy-based model | | ||
| | | | `s6_fig_pn.csv` | Cumulative success when screening viruses in the order recommended by the phylogenetic neighbourhood-based model | | ||
| | | | * | Note that this figure also displays values from Fig 1D (`fig1_d.csv`) for comparison | | ||
| | S7 Fig | | -- | Clustering based on data in S10 Fig, see plotting script linked below | | ||
| | S8 Fig | | `s8_fig.csv` | Clustering similarity comparisons using the Fowlkes-Mallows (Bk) index | | ||
| | S9 Fig | | `s9_fig.csv` | Membership of correlated-feature clusters and coordinates in 2 dimensions obtained via multi-dimensional scaling | | ||
| | S10 Fig | | `s10_fig.csv.zip` | SHAP values for each virus-genome feature combination, along with the underlying feature values | | ||
| | S11 Fig | | `s11_fig.csv` | Predictions for holdout viruses (also included in S1 Table) | | ||
| | S12 Fig | | -- | See data for S10 Fig | | ||
| | S13 Fig | | `s13_fig.csv` | AUC values for replicate training rounds with varying numbers of input features | | ||
|
|
||
|
|
||
| ## Interpretation | ||
| For the interpretation of these values, see the manuscript and the underlying code. All panels of a given figure are produced by the same script, as detailed below: | ||
|
|
||
| | Figure | Code | Command to reproduce figure and values<sup>[1](#footnote1)</sup> | | ||
| |---------|-------------------------------------------------------------------------------------------------------------|----------------------------------------| | ||
| | Fig 1 | [Scripts/Plotting/MakeFigure1.R](../Scripts/Plotting/MakeFigure1.R) | `make Plots/Figure1.pdf` | | ||
| | Fig 2 | [Scripts/Plotting/MakeFigure2.R](../Scripts/Plotting/MakeFigure2.R) | `make Plots/Figure2.pdf` | | ||
| | Fig 3 | [Scripts/Plotting/MakeFigure3.R](../Scripts/Plotting/MakeFigure3.R) | `make Plots/Figure3.pdf` | | ||
| | Fig 4 | [Scripts/Plotting/MakeFigure4.R](../Scripts/Plotting/MakeFigure4.R) | `make Plots/Figure4.pdf` | | ||
| | Fig 5 | [Scripts/Plotting/MakeFigure5.R](../Scripts/Plotting/MakeFigure5.R) | `make Plots/Figure5.pdf` | | ||
| | S1 Fig | [Scripts/Plotting/MakeSupplementaryFigure_RawData.R](../Scripts/Plotting/MakeSupplementaryFigure_RawData.R) | `make Plots/Supplement_RawData.pdf` | | ||
| | S2 Fig | [Scripts/Plotting/MakeSupplement_RelatednessModelRanks.R](../Scripts/Plotting/MakeSupplement_RelatednessModelRanks.R) | `make Plots/Supplement_RelatednessModelRanks.pdf` | | ||
| | S3 Fig | [Scripts/Plotting/MakeSupplement_FamilyAUC.R](../Scripts/Plotting/MakeSupplement_FamilyAUC.R) | `make Plots/Supplement_family_auc.pdf` | | ||
| | S4 Fig | [Scripts/Plotting/MakeSupplement_TrainingSetRanks.R](../Scripts/Plotting/MakeSupplement_TrainingSetRanks.R) | `make Plots/Supplement_TrainingSetRanks.pdf` | | ||
| | S5 Fig | [Scripts/Plotting/MakeSupplement_HighPriority_MissedZoonoses.R](../Scripts/Plotting/MakeSupplement_HighPriority_MissedZoonoses.R) | `make Plots/Supplement_HighPriority_MissingZoonoses.pdf` | | ||
| | S6 Fig | [Scripts/Plotting/MakeSupplement_ScreeningSuccessRate.R](../Scripts/Plotting/MakeSupplement_ScreeningSuccessRate.R) | `make Plots/Supplement_ScreeningSuccessRate.pdf` | | ||
| | S7 Fig | [Scripts/Plotting/MakeSupplementaryFigure_ClustersVsTaxonomy.R](../Scripts/Plotting/MakeSupplementaryFigure_ClustersVsTaxonomy.R) | `make Plots/Combine_tanglegrams.pdf` | | ||
| | S8 Fig | [Scripts/Plotting/MakeSupplementaryFigure_ClustersVsTaxonomy.R](../Scripts/Plotting/MakeSupplementaryFigure_ClustersVsTaxonomy.R) | `make Plots/Supplement_bk_plots.pdf` | | ||
| | S9 Fig | [Scripts/Plotting/MakeSupplementaryFigure_FeatureClusters.R](../Scripts/Plotting/MakeSupplementaryFigure_FeatureClusters.R) | `make Plots/SupplementaryFigure_FeatureClusters.pdf` | | ||
| | S10 Fig | [Scripts/Plotting/MakeSupplementaryFigure_EffectDirection.R](../Scripts/Plotting/MakeSupplementaryFigure_EffectDirection.R) | `make Plots/SupplementaryFigure_EffectDirection.pdf` | | ||
| | S11 Fig | [Scripts/Plotting/MakeSupplement_NovelVirusHosts.R](../Scripts/Plotting/MakeSupplement_NovelVirusHosts.R) | `make Plots/Supplement_NovelVirus_Hosts.pdf` | | ||
| | S12 Fig | [Scripts/Plotting/Supplement_IllustrateDerivedGenomeFeatureCalcs.R](../Scripts/Plotting/Supplement_IllustrateDerivedGenomeFeatureCalcs.R) | `make Plots/Supplement_methods_derived_genome_features.pdf` | | ||
| | S13 Fig | [Scripts/Plotting/MakeSupplement_FeatureSelection.R](../Scripts/Plotting/MakeSupplement_FeatureSelection.R) | `make Plots/Supplement_FeatureSelection.pdf` | | ||
|
|
||
|
|
||
| <sup name="footnote1">1</sup>See the [main README](../README.md#requirements) for setup requirements |
Oops, something went wrong.