The post Nvidia’s AI advances medical research to predict COVID-19 variant behaviors appeared on BitcoinEthereumNews.com.

Chip manufacturer Nvidia (NASDAQ: NVDA) has introduced an artificial intelligence (AI) model designed to generate gene sequences with close similarities to the virus behind COVID-19, SARS-CoV-2. In a blog post, the GenSLMs model can identify the differences between COVID-19 variants and classify genome sequences. Developed in partnership with academics from the Argonne National Laboratory and the University of Chicago, the large language model (LLM) was trained on a large data set of the nucleotide sequences—the foundations of DNA. Armed with the datasets, the researchers noted in their submission that the LLM demonstrated proficiency in generating sequences bearing close similarities to the COVID-19 strains that were prevalent in 2023. The researchers say GenSLMs were trained only with genomes from late 2019, with experts impressed by the ability of the model to predict virus mutations. “Our model’s generative process is extremely naive, lacking any specific information or constraints around what a new COVID variant should look like,” remarked GenSLMs lead researcher Arvind Ramanathan. “The AI’s ability to predict the kinds of gene mutations present in recent COVID strains—despite having only seen the Alpha and Beta variants during training—is a strong validation of its capabilities.” Apart from the functionalities of making seemingly accurate predictions, Nvidia says the model distinguishes between variants and interprets long strings of nucleotides. Using advanced software, researchers can visualize the evolutionary patterns of proteins in the COVID-19 genome, offering new use cases for medical research. “Looking at the model’s understanding of which mutations are particularly strong in a variant may help scientists with downstream tasks like determining how a specific strain can evade the human immune system.” said Ramanathan. GenSLMs is already carving a name in the ecosystem after research team members bagged the Gordon Bell special prize in 2022 at the SC22 supercomputing conference. The LLM relies on Nvidia A100 Tensor Core supercomputers,…

---

Filed under: News - @ November 20, 2023 6:24 pm