From Molecules to Medicines: Leveraging Artificial Intelligence for Next-Generation Drug Design

The integration of artificial intelligence (AI) into pharmaceutical research is rapidly transforming the drug discovery and development process. This study investigates the application of machine learning (ML) and deep learning (DL) algorithms in modern drug design, with a particular focus on identifying and optimizing novel bioactive compounds. We utilize curated datasets from reputable sources such as Drug Bank, Chambly, and PubChem, emphasizing molecules with established pharmacokinetic and pharmacodynamics profiles. Several models, including Random Forest, Support Vector Machines, Deep Neural Networks, and Graph Neural Networks, are trained to predict biological activity, ADMET properties, and drug-likeness of candidate molecules. The findings demonstrate that AI-driven models can significantly reduce the time and cost of drug development while enhancing prediction accuracy in early-stage screening. The study proposes a practical AI-based pipeline for identifying promising drug candidates, highlighting its potential to support more efficient and targeted pharmaceutical innovationsThe integration of artificial intelligence (AI) into pharmaceutical research is rapidly transforming the drug discovery and development process. This study investigates the application of machine learning (ML) and deep learning (DL) algorithms in modern drug design, with a particular focus on identifying and optimizing novel bioactive compounds. We utilize curated datasets from reputable sources such as Drug Bank, Chambly, and PubChem, emphasizing molecules with established pharmacokinetic and pharmacodynamics profiles. Several models, including Random Forest, Support Vector Machines, Deep Neural Networks, and Graph Neural Networks, are trained to predict biological activity, ADMET properties, and drug-likeness of candidate molecules. The findings demonstrate that AI-driven models can significantly reduce the time and cost of drug development while enhancing prediction accuracy in early-stage screening. The study proposes a practical AI-based pipeline for identifying promising drug candidates, highlighting its potential to support more efficient and targeted pharmaceutical innovations

کلیدواژه ها:

: Artificial Intelligence ، Drug Design ، Machine Learning ، Deep Learning ، Drug Discovery ، ADMET ، Bioactivity Prediction