Genomics Company, Amazon Partner on Generative A.I.

(Gerd Altmann, Pixabay. https://pixabay.com/illustrations/dna-analysis-research-7237234/)

20 Sept. 2023. A genomics company is providing precision medicine analytics from its databases with generative A.I. based on large language models from Amazon Web Services. Dante Genomics in New York says its collaboration with AWS will make clinical advice available to users, but also identify cases where a physician’s involvement is needed.

Dante Genomics is a seven year-old company that provides whole genome sequencing and analysis for individuals. As the name implies, whole genome sequencing queries the entire human genome, including regions not normally associated with disease or coding for proteins. Genes that produce instructions for proteins, called the exome, make up a small portion of the entire genome, and is the region most often sequenced and analyzed. As computer processing and data storage costs decline, however, sequencing the entire genome becomes more feasible, particularly when seeking answers for complex or rare conditions. Dante says whole genome sequencing results in 20 times more data than exome sequencing.

The company offers a series of standard reports from its analytics for common disease types such as cancer and heart disease, as well as rare diseases. Dante Genomics says it also produces customized reports that consider factors such as medical history, family history, and ethnic background. And the company says its reports can highlight latent carrier conditions and potential adverse effects to drugs not yet encountered. Dante also offers its software as a service called Clinical Lab in a Box for internal use by hospitals or clinics, as well as software offered over the web for one-time analysis of genomic data provided by the user.

Choose a model that best fits data sets

The collaboration with Amazon Web services aims to provide generative artificial intelligence from Dante’s whole genome data sets arrayed on the AWS Bedrock platform. The Bedrock platform, says AWS, provides large language models as a service through an application program interface or API. Large language models are machine learning algorithms trained by very large data sets such as text or image collections, and providing prospective outcomes, made famous recently by ChatGPT. In the Bedrock service, AWS says client companies can choose from a selection of foundation models that best conform to their data sets for machine learning, and test the pipelines — workflow processes for model and data generation — without investing in their own machine learning or A.I. infrastructure.

Dante Genomics says it plans to provide generative A.I. from the Bedrock platform, with customers interacting with the system through text chats. The company says it expects the system to produce genetic inferences to calculate the likelihood of disorders such as heart disease, respiratory conditions, or seizures. In addition, says Dante Genomics, the system will identify results where a physician should be consulted.

“By partnering with the tech experts at AWS,” says Dante Genomics co-founder and CEO Andrea Riposati in a company statement released through Cision, “Amazon Bedrock as a platform will help us make great strides in revolutionizing the everyday application of genomic data in personalized medicine, delivering better health outcomes with genomic data as a foundation.”

More from Science & Enterprise:

• Fujitsu, Linux Foundation Partner on Open-Source A.I.
• NIH Challenge Seeks Early Alzheimer’s Indicators with A.I.
• Biotech, A.I. Companies Model Animal Studies/Clinical Trials
• A.I. Analytics Detect Heart Failure in Single Ultrasound View
• Primate Genome A.I. Model IDs Human Disease Variants

We designed Science & Enterprise for busy readers including investors, researchers, entrepreneurs, and students. Except for a narrow cookies and privacy strip for first-time visitors, we have no pop-ups blocking the entire page, nor distracting animated GIF graphics. If you want to subscribe for daily email alerts, you can do that here, or find the link in the upper left-hand corner of the desktop page. The site is free, with no paywall. But, of course, donations are gratefully accepted.

*     *     *