This repository contains scripts for statistical inference and deep learning to study genetic interactions.
Since datasets are limited to very few type of cancer cell line models, we use an auto-encoder structure to learn the genetic background of cancer cells using gene expression data from the Cancer Cell Line Encyclopedia (CCLE) as part of DepMap project. Theoretically, the encoder part of the model should learn the genetic background of the cell lines represented in a low dimensional space. We then aim to use this representation of the cancer cell lines to enrich other models by adding this background information to the model. These models are those that predict genetic interaction between genes using experimental data in large scale combinatorial perturbation screen studies such as:
Horlbeck, et al. Mapping the Genetic Landscape of Human Cells. Cell (2018) [Paper] [Original Codes]