Pair wise drug to drug interaction

: pair wise synergy score prediction using Graph Neural Network

Abstract

Understanding the impact of synergistic (positive) or antagonistic (negative) pharmaceutical interactions on the efficacy of combination therapies is of paramount importance, given that several medical conditions need the administration of multiple medications concurrently. It is difficult to precisely identify and count the number of pharmaceutical interactions that really occur. The frequency of pharmaceutical interactions is contingent upon the kind of interactions included in research since many interactions may lack clinical relevance or be solely supported by theoretical evidence. Although it is probable that some individuals may have an unfavorable occurrence or side effect when using a combination of medications with known interactions, it is crucial to differentiate between possible drug-drug interactions (DDIs) and those that have significant therapeutic implications. The existing challenge of lacking a standardized method for quantifying synergy in chemotherapeutic drug combinations necessitates an exploration of diverse synergy metrics to formulate effective predictive models. Recognizing this gap, this study introduces a novel deep learning model that is designed to predict synergy scores for pair wise drug interaction. We propose deep learning-based model and, more specifically, network-based methods like graph neural networks (GNNs) which effectively predicts the synergy score that describes effect of the interaction of two drugs. By fusing multimodal input data such as gene expression of cancer cells and the chemical features of drugs, the model helps in predicting pair wise interaction scores. The GNN model demonstrates potential to predict Pair wise drug interactions. This study compares the prediction with Syn-Predict and achieves up to a 68% reduction in mean square error for Bliss and competitive results for HSA and CSS Synergy Scores.

Date of Award	2023
Original language	English
Awarding Institution	Western Sydney University
Supervisor	Yi Guo (Supervisor)