Trained on genome sequences rather than human language, PrimateAI-3D uses deep neural network architectures like ChatGPT and AlphaFold to help deliver personalised genomic medicine
Illumina launches PrimateAI-3D software. (Credit: PR Newswire/ Illumina, Inc.)
DNA sequencing company Illumina has launched the new PrimateAI-3D, an artificial intelligence (AI) software for accurate prediction of disease-causing genetic mutations in patients.
PrimateAI-3D uses deep neural network architectures like ChatGPT and AlphaFold to help deliver personalised genomic medicine.
According to Illumina, the algorithm is, however, trained on genome sequences rather than human language.
It uses a natural selection of the human genome to train the parameters of the deep neural network, using millions of benign genetic variants found in the sequencing of 233 diverse primate species.
Illumina claimed that this results in a deep neural network that can identify disease-causing variants with higher accuracy in all six clinical cohorts examined. Additionally, it has been shown to generate personalised predictions of genetic disease risk that have been verified in a cohort of almost half a million individuals.
Illumina CEO Francis deSouza said: “It is exciting to see PrimateAI-3D and the latest in AI technology combined with the most advanced DNA sequencing capabilities.
“Helping clinicians and researchers keep up with the vast quantities of genomic data now being generated from our platforms holds the potential to exponentially accelerate the critical work underway to better serve patients.”
In the next phase, Illumina scientists and academic collaborators applied the PrimateAI-3D to identify rare pathogenic mutations in approximately half a million people in the UK Biobank.
As per the findings, published in two papers in Science, the genomes of 97% of healthy individuals of the general population harbour highly actionable variants for a minimum of one of the 90 clinical conditions.
They also found that the software enhanced the accuracy of genetic risk prediction, the DNA sequencing firm added. This enables the first demonstration of polygenic risk scores that were largely unaffected by ancestry bias.
Illumina chief technology officer Alex Aravanis said: “The application of the latest advances in AI to genomics opens tremendous opportunities for Illumina in both genetic risk prediction and drug target discovery by decoding the basis of complex genetic diseases such as diabetes, heart disease, and autoimmune diseases.”