popGenMachineLearningExamples

This repository is meant to house a series of jupyter notebooks that showcase some simple examples of using supervised machine learning for population genetics inference.

The first notebook that we have added demographicModelSelectionExample.ipynb is meant as a companion to our recent review-- Schrider and Kern (2017) "Machine learning for population genetics: a new paradigm." That paper can be found on bioRxiv here: https://www.biorxiv.org/content/early/2017/10/20/206482

The subsequent notebooks really build on ideas presented in the first one. We aim to present a diversity of applications within population genetics that highlight a number of algorithms and ML practices. We recommend going through these in something like the following order:

1. demographicModelSelectionExample.ipynb- using a Random Forest classifier for doing demographic model selection
2. sweepDetectionExample.ipynb - using a Support Vector Machine classifier for detecting & categorizing sweeps
3. ancPopSizeRegressionExample.ipynb - using Random Forest regression to estimate past population size