What is it about?
In this work, we develop a deep learning pipeline to screen chemical databases for plant-based alternative medicines to treat Rheumatoid Arthritis (RA). The primary target is the cytokine protein tumor necrosis factor-alpha, a critical player in RA's complex immunological landscape. Additionally, we use a range of computational techniques such as ADMET, molecular docking, molecular dynamics, essential dynamics, and binding free energy calculations to evaluate these treatments at the molecular level, ensuring that they are both safe and effective with the potential for future clinical validation.
Featured Image
Photo by Towfiqu barbhuiya on Unsplash
Why is it important?
Conventional treatments for RA are often expensive and associated with toxic side effects. This emphasizes the need for safer and more sustainable alternatives. Our findings highlight the potential of lesser-known plant-derived compounds, such as Veratramine, Imperialine, Gelsemine, and Jervine, in inhibiting the inflammatory actions of the tumor necrosis factor-alpha protein and alleviating the systemic effects of RA. Our approach, which leverages the power of deep learning, is validated through extensive computational analysis and lays the foundation for future experimental investigations of the above inhibitors. Beyond RA, this method demonstrates broad applicability for studying other diseases, proteins, and larger datasets, highlighting its potential in computational drug discovery.
Perspectives
RA is a complex autoimmune disorder affecting 0.5-1% of the global population. While no cure currently exists, effective management is essential to alleviate symptoms and slow disease progression. However, the high costs and significant side effects of conventional treatments create substantial barriers, particularly for patients in developing countries like Bangladesh. We believe natural, plant-derived compounds offer a promising path to safer, more affordable, and effective alternatives for RA treatment. However, conventional methods of screening such compounds proved to be particularly laborious, non-exhaustive, and ultimately ineffective. This realization led us to adopt a deep learning-based predictive model in our search for viable natural solutions, as the advent of deep learning has revolutionized the discovery process for novel therapeutics. Encouragingly, the top compounds identified by our developed model exhibited excellent performance in computational metrics, which now paves the way for further in-vitro and in-vivo studies.
Akid Ornob
Military Institute of Science and Technology
Read the Original
This page is a summary of: Deep learning based predictive modeling to screen natural compounds against TNF-alpha for the potential management of rheumatoid arthritis: Virtual screening to comprehensive in silico investigation, PLOS One, December 2024, PLOS,
DOI: 10.1371/journal.pone.0303954.
You can read the full text:
Contributors
The following have contributed to this page
