Abstract :

Predicting Drug Responses Is A Crucial Difficulty In Computational Personalized Medicine. A Large Number Of Machine Learning Methodologies, Particularly Those Grounded On Deep Studying, Had Been Proposed For This Activity. Although, Those Techniques Regularly Describe Pharmaceuticals As Strings, Which Is Not A Natural Representation Of Molecules. Moreover, The Translation Of Factors Like As Mutations Or Replica Number Aberrations Influencing Remedy Response Has Now Not Been Comprehensively Addressed. Procedures: This Text Introduces A Unique Approach, Graph DRP, Utilizing A Graph Convolutional Network To Address The Hassle. In Graph DRP, Medicinal Drugs Were Shown As Molecular Graphs That Directly Represent Atomic Bonds, Even As Cell Traces Have Been Characterized As Binary Vectors Of Genomic Abnormalities. Convolutional Layers Learnt The Consultant Traits Of Medications And Cellular Traces, Which Have Been Sooner Or Later Incorporated To Symbolize Every Drug-cell Line Pair. The Reaction Fee For Each Drug-cell Line Pair Was Ultimately Predicted By A Fully Connected Neural Network. 4 Types Of Graph Convolutional Networks Were Employed To Analyze The Traits Of Prescribed Drugs. Findings: Graph DRP Surpasses TCNNS Throughout All Performance Metrics In Every Trial Performed. In Conclusion, Modeling Prescribed Drugs As Graphs Can Enhance The Efficacy Of Drugs Response Prediction. “Index Terms - Graph Convolutional Networks, Drug Response Prediction, Molecular Graphs, Genomic Aberrations, Machine Learning, Sentiment Analysis, Drug Reviews”.

Published:

09-6-2025

Issue:

Vol. 25 No. 6 (2025)

Page Nos:

42-50

Section:

Articles

License:

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

How to Cite