Deep Neural Networks in the Discovery of Novel Antibiotics Drug Molecule: A Review

International Journal of Research and Innovation in Applied Science (IJRIAS) | Volume V, Issue IX, September 2020 | ISSN 2454–6186

Deep Neural Networks in the Discovery of Novel Antibiotics Drug Molecule: A Review

Michael Idowu Oladunjoye¹, Olumide Olayinka Obe^{2
1, 2}Department of Computer Science, Federal University of Technology, Akure, Nigeria

Abstract – Machine learning methods have been used in various fields to enhance automation and the ability of computer to learn from experience for decades. Its application in drug discovery is emerging and is now being integrated into the process especially in the early stage for lead compounds screening. The antibiotic discovery process is cumbersome considering the duration and inadvertently the resultant cost. So, the application of deep neural networks will help in efficiency by reducing the duration of process and the overall cost of the process. A deep convolutional neural network successfully predicted new broad-spectrum antibiotic, Halicin, with other molecules with distinct structures. The deep learning models have been engaged to learn from known useful chemical compounds (molecules) with their biological activities. The use of the artificial intelligence techniques in the antibiotics discovery will be reviewed with focus on the deep neural networks model as compared with other methods.
Keywords – Machine learning, drug discovery, molecules, artificial intelligence, Deep neural networks
I. INTRODUCTION
Drug discovery is finding a promising molecule known as lead compound that could become a drug [2, 7]. It is also considered as the process of designing molecules that could someday lead to new therapies as part of drug development [3]. Generally, it is a process by which a drug candidate is identified before it is validated having been subjected to series of pre-clinical and clinical trials for the treatment of a specific disease. Before the discovery, efforts must be made to understand the disease to be treated and the underlying cause of the condition by scientist [2].
The first natural antibiotic product, Penicillin, was discovered by Alexander Fleming on the observation of a diffusible extract that had antibacterial activity against staphylococci produced by the Penicillium molds [4]. Antibiotic that can be effective against a particular biological target causing diseases involves reasonable numbers of experiments through a reasonable numbers of years to get the best (lead) potential molecules and this has been considered extremely cumbersome with advertently high cost [3]. The traditional approach to the discovery of new medicine involves screening of large number of compounds to identify a potential candidate and then the synthesis to optimize the molecular compound [6].

Read More