Mitigating cold-start problems in drug-target affinity prediction with interaction knowledge transferring

Abstract

Predicting the drug-target interaction is crucial for drug discovery as well as drug repurposing. Machine learning is commonly used in drug-target affinity (DTA) problem. However, the machine learning model faces the cold-start problem where the model performance drops when predicting the interaction of a novel drug or target. Previous works try to solve the cold start problem by learning the drug or target representation using unsupervised learning. While the drug or target representation can be learned in an unsupervised manner, it still lacks the interaction information, which is critical in drug-target interaction. To incorporate the interaction information into the drug and protein interaction, we proposed using transfer learning from chemical–chemical interaction (CCI) and protein–protein interaction (PPI) task to drug-target interaction task. The representation learned by CCI and PPI tasks can be transferred smoothly to the DTA task due to the similar nature of the tasks. The result on the DTA datasets shows that our proposed method has advantages compared to other pre-training methods in the DTA task.

BiComp-DTA: Drug-target binding affinity prediction through complementary biological-related and compression-based featurization approach

“…To evaluate the method’s robustness and generalization, especially for unseen data, we conducted widely-used alternative splitting data settings, named cold-start settings [ 24 , 44 ]. For this purpose, three settings have been applied for training and testing the method, including cold-protein, cold-drug, and cold-drug-protein for which, the model testing is performed for unseen protein, unseen drug, and unseen drug-protein pairs in the training set, respectively.…”

MULGA, a unified multi-view graph autoencoder-based approach for identifying drug–protein interaction and drug repositioning

“…Cold start poses a challenge and obstacle in drug repositioning ( Nguyen et al 2022 ). Since most proteins lack the interaction knowledge with drugs, we performed the “cold-start-for-protein” experiment to reposition the existing drugs as therapeutic targets of novel proteins.…”

MDTips: a multimodal-data-based drug–target interaction prediction system fusing knowledge, gene expression profile, and structural data

“…First, large amounts of known drug–target pairs (DTPs) must be used to train DL-based DTI models, while the labeled data volume is always limited. Second, they face cold-start problems, where the model accuracy decreases when predicting the interaction of a novel drug without knowing any target information ( Nguyen et al 2022 ).…”

“…Multimodal fusion, which has boosted the performance of many classical problems (e.g. visual question-answering) ( Xue and Marculescu 2022 ), is employed to integrate heterogeneous information from networks and automatically extract features of drugs and targets to facilitate further DTI prediction ( Nguyen et al 2022 ). Many models integrate diverse entities (e.g.…”