Machine learning pipeline to train toxicity prediction model of FunTox-Networks

Machine Learning pipeline used to provide toxicity prediction in FunTox-Networks

01_DATA # preprocessing and filtering of raw activity data from ChEMBL
- Chembl_v25    # latest activity assay data set from ChEMBL (retrieved Nov 2019)
- filt_stats.R  # Filtering and preparation of raw data
- Filtered      # output data sets from filt_stats.R
- toxicity_direction.csv    # table of toxicity measurements and their proportionality to toxicity

02_MolDesc # Calculation of molecular descriptors for all compounds within the filtered ChEMBL data set
- datastore     # files with all compounds and their calculated molecular descriptors based on SMILES
- scripts
    - calc_molDesc.py   # calculates for all compounds based on their smiles the molecular descriptors
    - chemopy-1.1       # used python package for descriptor calculation as decsribed in: https://doi.org/10.1093/bioinformatics/btt105

03_Averages # Calculation of moving averages for levels and organisms as required for calculation of Z-scores
- datastore     # output files with statistics calculated by make_Z.R
- scripts
    -make_Z.R   # script to calculate statistics to calculate Z-scores as used by the regression models
    
04_ZScores # Calculation of Z-scores and preparation of table to fit regression models
- datastore     # Z-normalized activity data and molecular descriptors in the form as used for fitting regression models
- scripts
    -calc_Ztable.py # based on activity data, molecular descriptors and Z-statistics, the learning data is calculated

05_Regression # Performing regression. Preparation of data by removing of outliers based on a linear regression model. Learning of random forest regression models. Validation of learning process by cross validation and tuning of hyperparameters.

- datastore     # storage of all random forest regression models and average level of Z output value per level and organism (zexp_*.tsv)
- scripts
    - data_preperation.R    # set up of regression data set, removal of outliers and optional removal of fields and descriptors
    - Rforest_CV.R          # analysis of machine learning by cross validation, importance of regression variables and tuning of hyperparameters (number of trees, split of variables)
    - Rforest.R          # based on analysis of Rforest_CV.R learning of final models

rregrs_output
# early analysis of regression model performance with the package RRegrs as described in: https://doi.org/10.1186/s13321-015-0094-2