Ai-Driven Covid-19 Mrna Vaccine Degradation Prediction

ABSTRACT

Background/Objective: Messenger RNA (mRNA) vaccines have emerged as a promising treatment for the coronavirus disease-2019 (COVID-19) pandemic, but such a solution has its challenges. One such issue is the mRNA vaccine's molecular stability, which requires that it be kept under certain environmental conditions that restrict its global outreach in packages, such as disposable syringes, distributed worldwide using refrigeration. Designing an environmentally stable mRNA vaccine that can withstand shipment worldwide is a challenge, since a single slit or puncture can render the complete dose of the vaccine useless. If not kept under certain environmental conditions and left unmonitored, mRNA vaccines tend to degrade rapidly. To address this problem, agencies currently store mRNA vaccines under strict refrigeration, thus limiting their global reach. The objective was to develop a hybrid deep learning model that can efficiently predict mRNA vaccine degradation rate from RNA sequences, thus aiding researchers and scientists in designing and developing a more stable mRNA vaccine in the future Results: Research presented here discusses the capability of the in-house developed hybrid deep learning model. Conclusion(s) The model was developed with a performance of 0.2430 mean columnwise root-mean-squared error (MCRMSE) score on the test data.