Synthetic lethality-mediated precision oncology via the tumor transcriptome
Creators
Description
We provide the source codes that were used to generate the results of our manuscript, ‘Synthetic lethality-mediated precision oncology via the tumor transcriptome.’ This software is available for academic use through a data sharing agreement.
1. Structure of the codes
The repository includes source codes under ./R directory and relevant data under ./data directory. To make predictions for treatment outcome, we used prediction.targeted.R for cytotoxic/targeted agents, and prediction.immuno.R for immune checkpoint therapies. We have a total of 23 datasets available in this repository with 10 cytotoxic/targeted agent cohorts in ./data/targeted directory and 13 immune checkpoint therapy cohorts in ./data/immuno. Not all the datasets used in the manuscript are made available here because some of the datasets are restricted to be shared due to data sharing agreements with the provider.
2. Platform used to test the codes
The codes were tested with R version 3.6.1 (2019-07-05) on a x86_64-pc-linux-gnu (64-bit) platform using R libraries ROCR v1.0-7, caTools v1.18.0, survival v3.1-11, survminer v0.4.6, Rcpp v1.0.3, and data.table v1.12.8.
3. How to install the codes
(1) Create a new directory named 'SELECT' and another directory named 'R' under 'SELECT' directory.
(2) Download the source code under 'SELECT/R' directory.
(3) Download the data under 'SELECT' directory. All the relevant data is compressed in https://drive.google.com/file/d/1hMK9d4icJHeQHu2BvbWl0umtwJihK2Fh/view?usp=sharing. Once extracted it generates ./data folder, and the data for inference of the SL/SR partners and test the predictions in clinical trial datasets are available.
cd SELECT
wget --no-check-certificate 'https://docs.google.com/uc?export=download&id=1hMK9d4icJHeQHu2BvbWl0umtwJihK2Fh' -O data.zip
(or type this link in your browser: ‘https://drive.google.com/file/d/1hMK9d4icJHeQHu2BvbWl0umtwJihK2Fh/view?usp=sharing’)
unzip data.zip
(4) Install R libraries as needed.
> install.packages("data.table")
> …
(5) Launch the code of interest
> source("./R/prediction.targeted.R")
> source("./R/prediction.immuno.R")