Language models improve vaccine development: #innovation.

Princeton University researchers have developed a groundbreaking language model that can leverage semantic representation to design enhanced mRNA vaccines, offering a promising avenue for developing more potent vaccines, including those crucial for combating COVID-19. The central dogma of biology explains the transfer of genetic information from DNA to RNA to proteins, with mRNA playing a key role in protein synthesis. The researchers focused on the untranslated region of mRNA to enhance vaccine efficiency and effectiveness. By training their model on a limited range of species and generating optimal sequences, they were able to improve protein production efficiency by 33%. This improvement could have a significant impact on the development of novel therapies for a wide range of infectious illnesses and malignancies. The model was trained on a relatively small number of sequences and included supplementary information regarding protein manufacturing. It generated 211 novel sequences aimed at enhancing translation efficiency, particularly for proteins like the spike protein targeted by COVID-19 vaccines. This language model is the first of its kind designed specifically for the untranslated region of mRNA and has the potential to predict a sequence’s performance in various related tasks.

Source link

Source link: https://indiaai.gov.in/article/unlocking-the-genetic-code-language-models-make-better-vaccines