A Deep Dive into Machine Learning: The Roles of Neural Networks and Random Forests in QSPR Analysis

Machine learning has significantly improved the field of drug development by enabling the accurate prediction of physicochemical properties and biological activities of compounds. Using machine learning and topological indices to analyze a drug’s structures can make process faster and more accurate....

Full description

Saved in:
Bibliographic Details
Published inBioNanoScience Vol. 15; no. 1
Main Authors Ahmed, Wakeel, Ashraf, Tamseela, AlMutairi, Dalal, Zaman, Shahid, Ahmed, Shakeel, Ehsan, Huma
Format Journal Article
LanguageEnglish
Published New York Springer US 01.03.2025
Springer Nature B.V
Subjects
Adaptive algorithms
Algorithms
Artificial neural networks
Biological activity
Biological and Medical Physics
Biological properties
Biomaterials
Biophysics
Circuits and Systems
Drug development
Drugs
Engineering
Learning algorithms
Machine learning
Nanotechnology
Neural networks
Physicochemical properties
Physiochemistry
Predictions
Topology
Random forest
Python algorithm
Topological indices
Machine learning
Artificial neural networks
Online AccessGet full text
ISSN2191-1630
2191-1649
DOI10.1007/s12668-024-01710-8

Cover

More Information
Summary:Machine learning has significantly improved the field of drug development by enabling the accurate prediction of physicochemical properties and biological activities of compounds. Using machine learning and topological indices to analyze a drug’s structures can make process faster and more accurate. Our study explores the molecular characteristics of 15 sulfur-based drugs ( S VI ) . Topological indices of these drugs have been calculated, and physiochemical properties have been examined using machine learning algorithms. Machine learning algorithms such as artificial neural networks, random forests, and adaptive boosting play a crucial role in this process. These algorithms utilize labeled data to make predictions about intricate molecular activities by assisting in the discovery of novel medication candidates and the enhancement of their properties. These algorithms enhance the accuracy of predictions related to physiochemical properties, reduce the time and cost associated with drug discovery, and rapidly analyze vast datasets by utilizing machine learning, consequently expediting the advancement of novel and efficient therapies.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2191-1630
2191-1649
DOI:10.1007/s12668-024-01710-8