Predicting Drug-Target Interactions and Recommending Targeted Therapies Using Multi-omics Data

Abstract

Drug discovery and development process is essential for addressing unmet medical needs and advancing personalized medicine. A critical aspect of this process involves identifying drug-target interactions (DTIs), which can inform the design of effective treatments. However, traditional laboratory-based DTI discovery methods are generally expensive and time-consuming. Therefore, computational approaches have been developed to predict interactions between drug candidates and disease-causing proteins. We designed a novel method, termed heterogeneous information integration for DTI prediction (HIDTI), based on the concept of predicting vectors for all of unknown/unavailable heterogeneous drug- and protein-related information. We applied a residual network in HIDTI to extract features of such heterogeneous information for predicting DTIs, and tested the model using drug-based 10-fold cross-validation to examine the prediction performance for unseen drugs. As a result, HIDTI outperformed existing models using heterogeneous information, was demonstrating that our method predicted heterogeneous information on unseen data better than other models. In conclusion, our study suggests that HIDTI has the potential to advance the field of drug development by accurately predicting the targets of new drugs. This research was published in Scientific reports, and the source codes are available https://github.com/DMCB-GIST/HIDTI.

Another important aspect of the drug discovery and development process is the emergence of targeted therapies. These therapies focus on identifying disease-related genes in each sample or patient and recommending drugs associated with those genes to enhance efficacy and reduce side effects. In our study, we used multi-omics data to identify potential driver genes specific to three sarcoma cell lines, SJCRH30, RD, and GIITC. Subsequently, we checked the potential of these genes as targets for drug candidates. We then predicted the affinities between FDA approved drugs and specific genes, and selected drugs with high affinity for each cell line. As a result, we were able to discover sarcoma-related genes and therapeutics that are effective against these cell lines. In conclusion, our pipeline, which recommends drugs targeting specific genes per cell line, has the potential to contribute to the development of personalized medicine, ultimately enabling tailored treatments for individual patients and improving outcomes.