Skip to content

Latest commit

 

History

History
29 lines (18 loc) · 1.29 KB

README.md

File metadata and controls

29 lines (18 loc) · 1.29 KB

malariaML

Authors: Nasib Mansour, Lydia San George, James Simmons

The purose of this repository is to develop and effective machine learning (ML) algorithm based off Cai John's Extreme Phenotype Sampling Machine Learning (ESL-ML) respository in order to predict clearance* for malaria when given dosages of pharmaceuticals. Additionally, this repository aims to assist clinicians, biologist, and other clinical or scientific field experts by providing a useful ML tool and to increase exposure to such methods.

The randomforest.py file analyzes the malaria data in order to predict clearence.

The ESL-ML_archieve contains the R files, which we base our analysis off of, comes from the ESL-ML repository.

The ESL-ML can be found here: https://github.com/caiwjohn/EPS-ML

Malaria Data

The data used comes from ESP-ML and each dataset originates from Zhu et al. and Mok et al. respectfully. The datasets are included in the malaria_data directory.

In order to effectily run a random forest algorithm, the original malarai date: mok_meta.txt and zhu_meta.txt is converted into .csv files.

Dependencies

Python3

Scikit-Learn

Build Procedure

python3 rf_mok.py

python3 rf_zhu.py

Appendix

Clearence - measurement of per volume of plasma in which a substance is completely removed per unit time.