Generating Labeled Genetic Synthetic Data to Non-Genetic Diseases
Creating large datasets for genetically influenced diseases is a hard and expensive task. We intend to use the idea behind Generative Adversarial Networks to artificially create cohesive labeled genetic data to characterize genetically influenced diseases, specifically Dengue.
Paper: A Semi-Supervised Generative Adversarial Network for Prediction of Genetic Disease Outcomes. Guide to the paper implementation here.
The branch tamu_hprc is meant to run on the High Performance Research Computing (HPRC) @Texas A&M University. Some environment configs are currently hard-coded on the code, so you'll need to make some changes to run it (I plan to move all of those to a new .env file as soon as possible). Also, this branch contains nightly modifications, which may be highly unstable at some points.