AI-Enhanced Drug Discovery Accelerating the Identification of Potential Therapeutic Compounds
Ram Madunuri, Sai Manoj Yellepeddi, Chetan Sasidhar Ravi, Subrahmanyasarma Chitta, Venkata Sri Manoj Bonam, Vinay Kumar Reddy Vangoor · 2024
This paper aims to propose the Enhanced Artificial Intelligence (EAI) approach to improve the speed of the identification of possible therapeutic agents in the process of drug discovery. Pharmacophore based methods are proven to have certain drawbacks like very high cost, huge time consumption and low hit rates. To overcome with these challenges the proposed system employs the sophisticated technologies of machine learning and deep learning. The general approach entails vast data gathering, preprocessing, and efficient feature extraction techniques and the use of various machine learning models such as Support Vector Machines (SVM), Random Forest (RF), Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN). The main advantage of the presented approach is the ability to compare it with traditional ones and highlight the advantages in terms of data processing, higher accuracy and less time consumption, as well as predictive accuracy. The experimental results presented also establish CNN with an accuracy of $92 \%$, the effectiveness of computational methods and blowing superiority in the experimental verifications. Such study also implies that the proposed approach with EAI integration has a huge potential of cutting on the cost and time taken in the overall process of drug discovery. This paper also discusses the possibilities for application of improved AI methods and enlargement of the existing data sets to improve the functionality of the system in the future.