Jiyin Lai

Abstract Tumor heterogeneity and drug resistance limit single-agent therapies, making combination treatments essential. However, traditional screening methods are costly and inefficient. Here, we present DSimSynergy, a graph deep learning framework for predicting drug synergy. It constructs drug similarity networks from biological process and clinical applications, then learns drug representations through graph convolution on these networks.

1 Introduction Cancer is a highly heterogeneous disease, displaying a range of genetic diversity and phenotypic variability.[1] Annually, numerous novel treatment modalities are subjected to clinical evaluation for their efficacy against diverse cancer types, yet less than 4% receive approval from the US Food and Drug Administration.[2] Even with favorable outcomes post-therapy, the considerable heterogeneity may lead to eventual tumor progression.[3] These challenges highlight the need for...

1 Introduction The occurrence and development of tumors are not an isolated event; rather, they develop within a complex microenvironment comprising various components and cell types.