Unsupervised-Decision-Trees

Data and code to support our paper titled: "Decision Tree Ensembles Utilizing Multivariate Splits Are Effective at Investigating Beta-Diversity in Medically Relevant 16S Amplicon Sequencing Data" https://www.biorxiv.org/content/10.1101/2022.03.31.486647v2

Requirements

To use the workflow LANDMark and TreeOrdination must be installed:

    LANDMark: https://github.com/jrudar/LANDMark
    TreeOrdination: https://github.com/jrudar/TreeOrdination

Input

To run the code, you will need the following files produced by the MetaWorks (https://github.com/terrimporter/MetaWorks) pipeline: 
    1) The ASV table (ESV.table)
    2) The Taxonomic Ranks and Confidence Scores of Each ASV (rdp.out.tmp)
    
In addition you will need the file containing the metadata.

These files are provided in this repository.

Also, please ensure that all paths are correct on your computer.        

Files

control_get_data.py:
    Description: Creates synthetic data and analyzes the results (Martino et al. 2019).
    
    Section 'run_exp': Calculates balanced accuracy scores and PerMANOVA results for each experiment.
    Section 'get_stats_perm': Plots the PerMANOVA results
    Section 'get_stats_bacc': Plots the Balanced Accuracy Score Results
    
diseased_gut_data.py:
    Description: Runs the experiment using the Crohn's Disease Data (Forbes et al. 2018).
    
    Section 'large_test': Calculates balanced accuracy scores and PerMANOVA results for each experiment
    Section 'feature_importance_test': Creates a TreeOrdination projection and calculates feature importance scores
    Section 'get_stats_perm': Plots the PerMANOVA results
    Section 'get_stats_bacc': Plots the Balanced Accuracy Score Results    
    
baxter_data.py:
    Description: Calculates ROC-AUC scores and Balanced Accuracy Scores using the Baxter et al. (2016) Data
    
baxter_data_fi.py:
    Description: Runs the experiment using the Baxter et al. (2016) data which calculates the sensitivity,
                 specificity, for the Normal vs. Cancer and Normal vs. Lesion comparisons. Calculates
                 Shapley scores.
                 
    Section 'get_stats': Plots ROC-AUC and Balanced Accuracy Scores
    Section 'normal_v_cancer': Calculates and saves CSV file of probability of cancer for each sample
    Section 'normal_v_cancer_stat': Calculates discrimination threshold, sensitivty, and specificity for
                                    the "normal_v_cancer" comparison
    Section 'normal_v_lesion': Calculates and saves CSV file of probability of lesion for each sample
    Section 'normal_v_lesion_stat': Calculates discrimination threshold, sensitivty, and specificity for
                                    the "normal_v_lesion" comparison
    Section 'get_fi_scores': Creates an TreeOrdination projection and calculates feature importance scores

Create the Environment

conda create -n TreeOrdFinal python=3.10

conda activate TreeOrdFinal
    
git clone https://github.com/jrudar/LANDMark
cd LANDMark
python setup.py sdist
cd dist
pip install LANDMark-1.2.0.tar.gz
    
pip install cython
cd ..
cd ..
git clone https://github.com/jrudar/TreeOrdination
cd TreeOrdination
python setup.py sdist
cd dist
pip install TreeOrdination-1.0.2.tar.gz
    
pip install statsmodels==0.13.5
pip install statannotations==0.5
pip install matplotlib==3.5.2
pip install seaborn==0.11.2

Data Sources and References

Forbes JD, Chen C-Y, Knox NC, Marrie R-A, El-Gabalawy H, de Kievit T, et al. A comparative study of the gut microbiota in immune-mediated inflammatory diseases-does a common dysbiosis exist? Microbiome. 2018 Dec 13;6(1):221–221.

Baxter NT, Ruffin MT, Rogers MAM, Schloss PD. Microbiota-based model improves the sensitivity of fecal immunochemical test for detecting colonic lesions. Genome Medicine. 2016 Apr 6;8(1):37

Martino C, Morton JT, Marotz CA, Thompson LR, Tripathi A, Knight R, et al. A Novel Sparse Compositional Technique Reveals Microbial Perturbations. mSystems. 2019 Feb;4(1).

Porter, T. M., & Hajibabaei, M. (2022). MetaWorks: A flexible, scalable bioinformatic pipeline for high-throughput multi-marker biodiversity assessments. PLOS ONE, 17(9), e0274260. doi: 10.1371/journal.pone.0274260

Wang, Q., Garrity, G. M., Tiedje, J. M., & Cole, J. R. (2007). Naive Bayesian Classifier for Rapid Assignment of rRNA Sequences into the New Bacterial Taxonomy. Applied and Environmental Microbiology, 73(16), 5261–5267. doi:10.1128/AEM.00062-07

Rudar, J., Porter, T.M., Wright, M., Golding G.B., Hajibabaei, M. LANDMark: an ensemble approach to the supervised selection of biomarkers in high-throughput sequencing data. BMC Bioinformatics 23, 110 (2022). https://doi.org/10.1186/s12859-022-04631-z