This is the home repo for the Neural Network team in the DiscoveryDNA group. The aim is to build and employ a Neural Network, specifically a Recurrent Neural Network (RNN) for predicting enchancer function in non-coding regions of Drosophila genomes. We use Python, Keras, Tensorflow, and Amazon Web Services as our enviroment.
Current: Adam Stafford, Zhanyuan (Sean) Zhang, Jiaxi (Jake) Zhao, and Ciera Martinez
Previous: Tianhao Zhang,Yichen Fang, Boren Tsai, Thomas Lane,
If you plan on working on this team or contributing to this project, please read data_management.md document which details the standards we have set to work coherenltly as a team. Every exeriment and data cleaning step should be carefully documented in a Jupyter Notebook and should be created with reproducbility in mind. Everyone on the team (and future you) should be able to re-create what you have performed.
Our data is backed up on Google Drive. Our working data is on our AWS servers and will be fully released upon publication. If you are outside of our Team at Berkeley and would like to contribute to our project, please contact Ciera Martinez at ccmartinez@berkeley.edu.