Using graph-based model to identify cell specific synthetic lethal effects

Abstract

Synthetic lethal (SL) pairs are pairs of genes whose simultaneous loss-of-function results in cell death, while a damaging mutation of either gene alone does not affect the cells survival. This makes SL pairs attractive targets for precision cancer therapies, as targeting the unimpaired gene of the SL pair can selectively kill cancer cells that already harbor the impaired gene. Limited by the difficulty of finding true SL pairs, especially on specific cell types, the identification of SL targets still relies on expensive, time-consuming experimental approaches. In this work, we utilized various cell-line specific omics data to design a deep learning model for predicting SL pairs on particular cell-lines. By incorporating multiple types of cell-specific omics data with a self-attention module, we represent gene relationships as graphs. Our approach demonstrates the potential to facilitate the discovery of cell-specific SL targets for cancer therapeutics, providing a tool to unearth mechanisms underlying the origin of SL in cancer biology. Our approach allows for prediction of SL pairs in a cell-specific manner and enhances cancer precision medicine. The code and data of our approach can be found at https://github.com/promethiume/SLwise HighlightsO_LIFew computational methods can systematically predict SL pairs at a cell-specific level, and their performance may not generalize well to clinical scenarios due to the heterogeneity of cancer types. C_LIO_LIThe SLWise utilizes various cell-line specific omics data to design a deep learning model with a graph-based representation and self-attention mechanism. C_LIO_LIThis approach allows for the prediction of SL pairs in a cell-specific manner, providing valuable insights on effectively identifying the cell-type specific SL targets for personalized treatment strategies. C_LI