By combining the strengths of both parties in genomic interpretation and genomic analysis technologies, as well as industry expertise, to accelerate the discovery of drug targets. This collaboration will evaluate whether a comprehensive framework integrating these technologies can enhance the productivity and reliability of drug target discovery and develop promising drugs based on human omics knowledge. If the initial collaboration is successful, the two companies will reassess a long-term partnership and further advance the development of targets and drugs, clinical trials, and submissions to regulatory authorities.
This collaboration leverages Illumina's next-generation AI-based tools for interpreting genomic data, PrimateAI and SpliceAI, as well as AstraZeneca's analytical framework for rare variant genomic discovery, along with the latter’s own AI tools, including JARVIS and missense tolerance predictors among other computational predictors. PrimateAI is a deep neural network trained on hundreds of thousands of genetic variants collected from both humans and animals, capable of predicting which mutations may cause disease. SpliceAI is an independent open-source deep learning tool designed to uncover previously overlooked non-coding mutations that could be associated with rare genetic disorders and other diseases. JARVIS is a comprehensive deep learning framework used to scan non-coding regions across the entire genome to identify potentially harmful variants. AstraZeneca will integrate all these AI programs into a larger framework, which will then be used to analyze large-scale multi-omics datasets from its digital biobank.These AI tools, with complementary functions, will better identify pathogenic gene variants, providing a crucial foundation for the subsequent development of safe and effective therapies.
Illumina's Chief Strategy and Corporate Development Officer and Interim Chief Financial Officer, Joydeep Goswami, stated: "Illumina and AstraZeneca each possess industry-leading capabilities and will be in a unique position to enhance drug development efficiency by identifying pathogenic variants. By pinpointing genes that show evidence of a causal relationship with human diseases, the combined AI framework has the potential to prioritize candidate drugs that are more likely to gain approval."
Figure 1 Joydeep Goswami (Source: Illumina official website)
Slavé Petrovski, Head of AstraZeneca's Genomics Research, Discovery Science, and R&D Center, stated: "The continuous innovation of AI tools and frameworks applied to the growing resources of human genomics, transcriptomics, and proteomics medical research will enable us to address some previously challenging questions, help us discover new drug targets with higher success rates, and identify which patients are most likely to benefit from the new therapies we are developing."
Figure 2 Slavé Petrovski (Source: AstraZeneca official website)
This is not the first time Illumina has set its sights on the pharmaceutical field, it can be said that,Over the past year, Illumina has been making frequent moves, systematically venturing into the AI-driven drug discovery field. This marks a shift from its previous business model of selling sequencing equipment to various biotech companies, transitioning from being a "tool provider" to directly participating in the search for valuable breakthroughs.In January this year at the J.P. Morgan Healthcare Conference, Illumina announced a long-term agreement with Nashville Biosciences, a wholly-owned subsidiary of Vanderbilt University Medical Center (VUMC), to jointly unlock 250,000 whole genome sequences linked to clinical information from VUMC's biobank database, thereby advancing new methods for diagnosing, preventing, and treating a range of human diseases. On April 27, Illumina announced at the World Companion Diagnostics Summit a long-term strategic collaboration with Janssen Biotech, a subsidiary of Johnson & Johnson, utilizing Illumina’s portfolio of resources including genomic interpretation tools, whole genome sequencing, and multi-omics analysis services to co-develop companion diagnostics for cancer. In the same month, Illumina announced a five-year partnership with healthcare investment management firm Deerfield Management to use genomic tools and genetic knowledge to select projects with higher probabilities of success, reduce R&D costs, and accelerate the approval of cutting-edge therapies.
These collaborations will bring downstream revenue sharing to Illumina, Goswami said, which is part of Illumina's long-term strategy: "It takes more than 10 years for these things to come to market and be commercialized." But the prospect of lower drug development risks and shorter approval times for market entry is enticing. In the process of drug development, pharmaceutical companies often face an overwhelming amount of information but struggle to generate significant insights. "As genomics experts, we have a unique perspective. These AI tools and frameworks consolidate decades of our experience, allowing us to analyze genomic data and extract signals from the noise, offering suggestions to pharmaceutical companies in areas where their capabilities may be lacking."
Illumina's moves are not solely driven by the consideration of diversifying its business from sequencing tools to address potential market share loss. Facing the common challenges of long development cycles, high costs, and risks in new drug research and development, applying AI to target discovery, compound synthesis, new indication discovery, compound screening, crystal form prediction, patient recruitment, optimization of clinical trial design, and drug inspection undoubtedly represents a future direction for improving success rates and reducing costs. The solution to one of biology’s 50-year-old grand challenges—protein folding prediction—came from AlphaFold, a deep learning algorithm developed by DeepMind, an artificial intelligence company under Google. Illumina's Chief Technology Officer, Alex Aravanis, pointed out: “The intersection of human genetics and AI will bring about the next generation of drug discovery."As the dominant player in the gene sequencing market, Illumina happens to possess the tools to address the current unmet needs in disease treatment solutions."
However, AI-driven drug discovery is still in its infancy, and globally, no drug developed by AI has successfully reached the market. IT giants, AI drug discovery startups, and large pharmaceutical companies are entering the industry, leveraging their respective advantages in funding, computing power, and data experience. Whether they can find a suitable business collaboration model to complement each other's strengths and achieve profitability remains a question that only time can answer.
References:
[1]https://www.fiercebiotech.com/medtech/illumina-joins-astrazeneca-ai-driven-drug-discovery-project
[2]https://www.illumina.com.cn/company/news-center/press-releases/press-release-details.html?newsid=610e7fbe-cae9-4f64-a07d-8ebabce60b38
[3]https://www.liebertpub.com/doi/full/10.1089/genbio.2022.29036.gil
