In silico learning of tumor evolution through mutational time series

Citation

Noam Auslander, Yuri I Wolf, Eugene V Koonin. In silico learning of tumor evolution through mutational time series. Proceedings of the National Academy of Sciences, April 2019.

System requirements:

1. MATLAB has to be installed.
2. The following packages should be installed: a. Deep Learning Toolbox b. Statistics and Machine Learning Toolbox

Version checked on: MATLAB2018a,MATLAB2018b

OS tested: mac OS 10.12.6 and 10.10.5, NIH HPC linux cluster

This code has four parts

1. PART1_PREDICT_LOAD - predict mutational load from a time series of mutations
2. PART2_PREDICT_SEQ - predict the next mutation in the time-sequence
3. PART3_CONSTRUCT_DATA - construction of simulated mutational data
4. PART4_PREDICT_INTERACTIONS - predicting occurrence of mutations from the binary sequence of major drivers, validation of interactions and survival analysis

Each part contains a separate, detailed README file