Illumina unveils PromoterAI, a groundbreaking algorithm to accelerate insights for rare disease diagnosis

Original from: Illumina

Illumina Inc. (NASDAQ: ILMN) today unveiled PromoterAI, a new AI algorithm that accurately deciphers pathogenic regulatory genetic variants in the noncoding regions of the human genome. A study published today in Science illustrates how this deep learning technology discovered regulatory variants in noncoding "promoter" segments that contribute up to 6% of the genetic causes of rare diseases. By turning data into insights, this technology is expected to help researchers accelerate new breakthroughs in diagnosis for rare diseases.

Currently, only about 30% of rare disease patients receive an accurate diagnosis from exome sequencing. For many patients, insights may be hidden in what are known as "noncoding regions," which comprise over 98% of the genome and may hold the key to increasing the diagnostic rate.

Promoters are key noncoding elements that define where transcription of a gene begins. When promoters are disrupted by genetic mutations, it can have profound consequences for downstream gene expression. The PromoterAI algorithm interprets the genetic sequence of promoters, thus opening the door to precise diagnoses for more patients with rare diseases.

"Illumina is breaking new ground with this AI technology," said Kyle Farh, MD, PhD, vice president, distinguished scientist, and head of the Illumina Artificial Intelligence Lab. "Combined with whole-genome sequencing, we believe PromoterAI holds the power to improve the rate of rare disease diagnosis by identifying previously overlooked variants in the noncoding genome. One of the challenges of searching for the genetic causes of rare disease is that, even if the protein-coding sequence of a gene is free from variants, mutations in that gene's promoter region can prevent it from being properly expressed."

Built on state-of-the-art deep learning neural networks, PromoterAI brings unprecedented precision to the interpretation of promoter variants. With PromoterAI applied to whole-genome sequencing (WGS) data analysis, clinical researchers are empowered to better understand the etiology and drug target potential for rare genetic diseases, common diseases, and oncology.

PromoterAI is the latest genomic AI algorithmic innovation from Illumina. In 2019, the Illumina Artificial Intelligence Lab released SpliceAI, a deep learning tool for interpreting noncoding cryptic splice mutations. In 2023, Illumina released PrimateAI-3D, which predicts the pathogenicity of protein-coding variants based on evolutionary conservation and protein structure. The study demonstrates that when used together, Illumina's AI classification prediction tools—PromoterAI, PrimateAI-3D, and SpliceAI—effectively double the diagnostic yield compared to using protein-truncating variants alone. 

"PromoterAI is the latest addition to our AI and software portfolio, which has delivered leading quality, prediction accuracy, and comprehensiveness to turn data into valuable insights for our customers," said Rami Mehio, vice president, and global head of Software and Informatics.

PromoterAI will be accessible through Illumina Connected Software and it is available today as part of DRAGEN™ secondary analysis. Precomputed PromoterAI scores for all human promoter single-nucleotide variants are freely available for academic and noncommercial research use. Illumina Connected Software supports seamless integration with next-generation sequencing and array workflows for genomics and multiomics. Researchers, bioinformaticians, and clinical geneticists can use Illumina Connected Software to prioritize variants that are important to a case, gain critical insights, and accelerate their precision medicine dry lab operations.

Source: Illumina unveils PromoterAI, a groundbreaking algorithm to accelerate insights for rare disease diagnosis