Integrated Molecular Modeling and Machine Learning Protocol to Target Interleukin-6 (IL-6) and Tumor Necrosis Factor α (TNF-α) for Effective Treatment in Rheumatoid Arthritis

Abstract

As a complicated chronic autoimmune disease, Rheumatoid arthritis (RA) often causes progressive cartilage damage and grave inflammatory comorbidities in the joint tissues, substantially reducing the life quality and mortality in patients. The primary cornerstones of this study are the cytokines IL-6 and TNF-α as they are elevated in patients with RA and play a vital role in the pivotal administration of inflammatory responses in the disease process as well as being linked with polymorphism in genes incorporated in monitoring inflammatory passageways. Therefore, various reports suggest targeting both IL-6 and TNF signaling could be a propitious strategy to treat RA because of crucial role of TNF-α and Il-6 in its pathophysiology. The pathways to be targeted in this study are the (MAPK) signaling pathway and (JAK/STAT) signaling pathway which together form the Toll-like receptor

pathway that suppresses activation by generating pro-inflammatory mediators as well as self- regulating signals. Here, in this study, molecular modeling strategies along with supervised

machine learning models were implemented to predict important 2D and 3D anti- inflammatory features of TNFα and IL-6, respectively. To achieve this purpose, molecular

docking protocol optimization was performed to probe the binding hypothesis of both the targets and to validate the binding pattern and explore the stability of these inhibitors‟ complexes with the TNF-α, MD simulation was used. Moreover, machine learning classification models were used to identify the most relevant 2D features for both targets. Conjointly, this study elucidates the salient 2D features along with crucial interactions present in the protein targets that might assist in the development of more potent drugs for the treatment of this inflammatory disease.