Digital visualization of data flow and formation of neural networks, database processing

[Adobe Stock/max_776]

As cloud technology matures, life-sciences companies are finding new avenues for business value in life sciences, extending beyond the cost efficiencies typically associated with cloud migrations, as McKinsey noted. In an email statement, Microsoft outlined how its own cloud computing, AI and research capabilities are driving similar innovation for biopharma companies. This article explores Microsoft Research initiatives related to drug discovery, including technology developments and strategic collaborations.

Quantum Computing’s role in Microsoft Research’s drug discovery and scientific research push

A key facet of Microsoft Research’s impact on drug discovery lies in its quantum computing initiatives. In terms of quantum computing efforts, the company said, “We’d like to first share with you our recent announcement on new advances to Azure Quantum, aimed at accelerating scientific discovery. Azure Quantum Elements, Copilot in Azure Quantum and our progress to a quantum supercomputer are key elements to transforming drug discovery and development in the future.”

Emerging projects with drug discovery relevance

Additionally, through Microsoft Research, the company has a number of emerging projects and collaborations that contribute to advances in drug discovery. “Microsoft Research Health Futures is focused on making healthcare more data-driven, predictive and precise — ultimately, empowering every person on the planet to live a  healthier future,” the company said.

The company is also committed to advancing biomedical natural language processing (NLP).” A crucial factor in future-facing, data-driven health systems is the accessibility and interpretability of multimodal health information,” the company said. “To meet this need, Microsoft has laid a solid foundation across multiple modalities in biomedical NLP building on our deep research assets in deep learning and biomedical machine reading.”

Examples include the following:

  • AI4Science: A global organization of researchers and engineers dedicated to advancing  a new research frontier at the intersection of machine learning and the natural  sciences.
  • Graphormer: Developed by Microsoft Research, Graphormer is a deep learning  package that allows researchers and developers to train custom models for  molecule modeling tasks. It aims to accelerate the research and application in AI  for molecule science, such as material discovery and drug discovery.
  • Project Science Engine: Harnesses advances in AI and high-performance computing to accelerate research and development processes in industries that are anchored in hard sciences, including biopharma, chemicals, materials, cosmetics, foods, alternative  energy and decarbonization.

Case studies

Microsoft provided several examples of collaborations where their technology has accelerated drug discovery timelines:

  • The Microsoft Research project MoLeR demonstrated how AI could analyze molecular interactions to predict chemical structures with desired therapeutic properties faster. This enables more efficient drug design.
  • Working with Novartis, Microsoft AI helped speed the discovery and development of breakthrough medicines. By analyzing vast amounts of data, Microsoft technology enabled Novartis to better understand diseases and identify potential treatments.
  • In partnership with Novo Nordisk, Microsoft is developing a platform using AI and machine learning to accelerate drug discovery and development. The platform analyzes large volumes of data to derive insights and enable decisions.
  • Through an AI innovation partnership with UCB, Microsoft Research utilized AI technology to explore new disease targets and drug candidates. Advanced algorithms were applied to UCB’s data to uncover hidden patterns.
  • Microsoft partnered with Adaptive Biotechnologies to create a clinical test to aid in early detection of Lyme disease. Drawing on genetics and AI, the T-Detect test analyzes the immune response to diagnose Lyme disease weeks earlier than existing methods.

Microsoft partnership with OpenAI

Microsoft discussed the massive potential of generative AI models like OpenAI’s ChatGPT in life sciences, but noted that realizing the full possibilities requires a solid data foundation. “Without a more comprehensive data strategy built into their organization’s architecture, they’ll only scratch the surface of what is possible. AI  transformation relies on a solid journey to bring digital capabilities to the cloud across the  entire organization. “

Microsoft Research drug discovery HPC acceleration

Microsoft stated that Azure’s high-performance compute (HPC) capabilities have accelerated critical research into COVID-19 and other diseases: “Azure HPC has enabled researchers to accelerate insights and advances into genomics, precision medicine, and clinical trials with near-infinite high-performance bioinformatics infrastructure capabilities,” the company said.

Specifically, Microsoft provided examples of partnerships where Azure HPC was used:

  • Adaptive Biotechnologies partnered with Microsoft to create ImmuneCODE, an expansive map of the immune response to COVID-19. It was generated from thousands of de-identified blood samples analyzed using Azure HPC.
  • ImmunityBio tapped Azure HPC to computationally analyze the SARS-CoV-2 spike protein. This enabled them to create a detailed 3D model to inform vaccine and therapy development.
  • The Molecular Modelling Laboratory used over 18,000 Azure virtual machines to run high-throughput drug screening. This helped optimize their pipeline to alleviate common drug development hurdles.

The scalable computing power of Azure HPC has allowed researchers to rapidly analyze genomic, protein, and other biomedical data. This has accelerated insights in COVID-19 as well as other therapeutic areas. Azure provides researchers with on-demand access to cloud-based supercomputing capabilities.

A record-breaking simulation with Azure HPC

Microsoft discussed a recent record-breaking computational fluid dynamics (CFD) simulation enabled by Azure HPC:

“Late last year, researchers and technologists from Microsoft, Ansys, the University of Eindhoven in the Netherlands, and KU Leuven in Belgium conducted one of the largest CFD simulations using Azure HPC. The team tracked aerosol distribution among 30,000 spectators in a stadium through a 6 billion cell simulation, the largest ever performed with a commercially available CFD solver.”

This unprecedented simulation provided new insights into airborne virus transmission in crowded spaces, advancing research during the pandemic.

More broadly, the scalable computing power of Azure HPC empowers researchers across disciplines by providing on-demand access to advanced modeling, simulation, and analysis capabilities. The flexible infrastructure can be customized to research needs, enabling complex computations that accelerate discoveries.

By eliminating physical and cost constraints of traditional HPC resources, Azure allows researchers to harness previously unattainable computing power, storage, and analytics capabilities. This is accelerating research in a wide range of fields from disease modeling to weather forecasting to high-energy physics.

Strategic collaboration with Broad Institute, Harvard and Verily: Terra Platform

Microsoft discussed their collaboration with Broad Institute, Harvard and Verily on the Terra platform: “The strategic partnership between Microsoft, Broad Institute of MIT and Harvard, and Verily aims to accelerate new innovations in biomedicine through the Terra platform,” the company said. “Terra is a secure, scalable, open-source platform for biomedical researchers to access data, run analysis tools and collaborate.”

Terra provides researchers with a cloud-based platform to store, analyze, and share biomedical data. Microsoft Azure will provide the cloud infrastructure, services, and technologies to enhance Terra’s capabilities.

Specifically, Azure’s scalable computing power, ability to manage vast datasets, and AI tools like machine learning will augment what Terra can offer researchers. This will provide scientists with greater computing resources and more advanced analytics capabilities on the Terra platform.

Ultimately, Microsoft’s contribution aims to empower researchers and clinicians with improved access to biomedical data along with enhanced collaboration tools. This will help accelerate discoveries and innovation in areas like genomics, biotech, and precision medicine to improve patient outcomes.