Machine learning approaches to classify COPD stable and exacerbation stages

Caret package in R is used to bulid different machine learning models (Support Vector Machine, Random Forest, Naive bayes, Generalized Linear Model and Multilayer perceptron) using the ASV matrix from Amplicon-16S sequencing data of COPD stable and exacerbations from sputum samples retrieved from MDPD.

Dataset 1

It includes the ML input file (ASV matrix) and all the models for dataset 1. In this dataset stable, infrequent exacerberator and end of exacerbations are denotesd as COPD stable while exacerbation, start of exacerbation, frequent exacerberator are regarded as COPD exacerbation subgroup.

Dataset 2

It includes the ML input file (ASV matrix) and all the models for dataset 2. In this dataset only stable is denotesd as COPD stable while exacerbation, frequent exacerberator are regarded as COPD exacerbation subgroup.