Application of Artificial Intelligence and Computational Biology in the In Silico Design of a Tolerogenic Vaccine Against Myasthenia Gravis: A Novel Approach for Autoimmune Disease Therapy

In this study, artificial intelligence and computational biology approaches were employed for the in silico design of a tolerogenic vaccine targeting the autoimmune disease myasthenia gravis. CD۴⁺ T-cell epitopes derived from the α۱ subunit of the nicotinic acetylcholine receptor (nAChR α۱) were predicted and rigorously screened using AI-based tools to ensure favorable immunological properties, including non-allergenicity, non-toxicity, IL-۱۰ induction, and absence of IFN-γ induction. The selected epitopes were joined using optimized linkers and combined with human IL-۱۰ and the L۷/L۱۲ ribosomal protein as an adjuvant. Structural modeling was performed via ColabFold (based on AlphaFold۲), followed by atomic-level refinement using ModRefiner. Molecular docking and structural dynamics analysis demonstrated stable interactions between the vaccine construct and immune receptors (TLR۴ and IL-۱۰R), with favorable binding energies and structural stability. These findings underscore the potential of AI- and computational biology-driven vaccine design to produce safe and effective tolerogenic vaccines for autoimmune diseases such as myasthenia gravis. This in silico approach offers a promising framework for accelerating the rational development of next-generation vaccines targeting immune tolerance.

کلیدواژه ها:

Artificial Intelligence ، Computational Biology ، Tolerogenic Vaccine ، Autoimmune ، Myasthenia Gravis