Originally published Aug 31, 2020

by Elizabeth Cutler, Co-Founder, AI in Biopharma

As a Co-Founder of the AI Applications in Biopharma Summit and the AI in Biopharma Digital Summit Series, I am working to provide valuable information about AI applications to the biopharma community so that ultimately, treatments can get to patients faster.  Below you will find selected talking points that I found to be thought-provoking and useful insights from the virtual event STAT hosted in August 2020 on the topic of, Cracking Covid-19’s Code with AI.  Please note that these are summaries and not verbatim comments.

Topics discussed at the virtual event include

  • The Use of Big Data in Healthcare
  • Updates on Current Projects
  • Collaboration and Bottlenecks
  • The Realities of RWE and Standardization
  • The Full Spectrum of AI Technologies
  • Bringing Together Science and Technology
  • Final Thought
  • About the AI in Biopharma Digital Series

Participating speakers

Julie Rubinstein, President, Adaptive Biotechnologies
Karen Akinsanya, Ph.D., Executive Vice President, Chief Biomedical Scientist, Head of Discovery R&D, Schrödinger
Peter Lee, Corporate Vice President, Microsoft Research & Incubations
Matthew Might, Ph.D., Director, Hugh Kaul Precision Medicine Institute, University of Alabama of Birmingham
Casey Ross, National Technology Correspondent, STAT (moderator)

The Use of Big Data in Healthcare

Julie Rubinstein
Covid-19 has fueled a transformation of the use of big data in healthcare. Collaborative efforts, often propelled by technologies like AI, have accelerated our ability to obtain information about the virus and have fundamentally changed the way vaccines and therapeutics are being altered and developed so that patients can get the care they need

Updates on Current Projects

Matthew Might
Matthew has launched a clinical trial for treatment of severe Covid-19 cases using an AI tool that draws on many different biomedical data sets. The data is transformed into knowledge using logic-driven AI to focus on modulating the relevant targets for Covid-19.

Karen Akinsanya
Karen’s work is part of a consortium of pharmaceutical companies using AI to identify small molecules that could serve as an anti-viral. The consortium is applying their collective capabilities to be able to rapidly come up with anti-virals that will be the next wave of important treatments.

Peter Lee
Peter is working closely in partnership with Adaptive Biotechnologies and has focused on Adaptive’s unique ability to analyze the genetics of t-cells. The t-cell response sends very strong and early signals and, in some ways, can provide more information than simply analyzing fragments of the virus. Machine learning is used in three stages with the work on t-cells. This results in diagnosis capabilities but also provides the potential to show likely responses to treatments, should a person get infected. For vaccine and drug development there could be a further use for this data to show how people are responding to drug therapies.

Collaboration and Bottlenecks

Matthew Might
Covid-19 has been the ultimate bottleneck breaker in terms of collaboration. Partnerships that have formed during the pandemic have been able to get through bottlenecks that were previously unthinkable. This has resulted in ample data and ample compute. One bottleneck that still exists is that many of the interesting papers are still locked behind paywalls.

Karen Akinsanya
With a million papers being published every year, it’s really imperative that scientists have access to that information. One collaborative tool Schrödinger is using holds data in a central location and this has been very useful for increasing the ability to share data effectively. Interoperability between companies remains difficult but Covid-19 presents an immediate and urgent need to speed up a solution to interoperability problems to foster collaboration and speed going forward

Data Privacy, Access and Consent

Dan Vahdat
Patient’s data is owned by the patient. The patient has the option to decide if they want to share their data for clinical care or research purposes. Even sharing anonymized data, you need to be thoughtful about why you are sharing data and who you are sharing it with.

John Halamka
GDPR provides a guardrail, or set of guidance, about using data ethically but culture is even more important than GDPR. Pay attention to culture because sometimes that is the real limiting factor. Engaging patients is key to ensuring that AI can be adopted faster.

Angeli Moeller
Giving the power to the patient regarding what privacy and confidentiality they accept is critical.

The Realities of RWE and Standardization

Peter Lee
Pulling EHR data has begun to expose the realities of RWE and the difficulties of combining genomic data with clinical records and applying machine learning. However, the learnings coming from doing this work during Covid-19 have focused attention on the important problems to solve, including productive discussion about standardization.

The Full Spectrum of AI Technologies

Peter Lee
When talking about AI, we are generally talking about machine learning. Within machine learning there are deep learning techniques that provide a boosting process. Starting with a small cohort providing data can then be boosted using AI tools, even if most of the data is naïve data. The other great use of AI tools is the ability to sort through all of the literature available that could be useful to solve a particular problem. Auto-ML techniques are just starting to get deployed but show promise.

Matthew Might
Matthew uses a full spectrum of AI technologies with a lot of the portfolio being focused on old fashioned logic-driven AI. This is useful for deducing non-obvious facts from literature by combining disparate pieces of knowledge together.

Karen Akinsanya
Looking through existing data sets, in particular compounds, can be very powerfully impacted by machine learning to extract key pieces of information. When thinking about the use of deep learning, when looking for molecules in chemical space that may not have found their way into existing libraries, Schrödinger relies first on creating an understanding of molecular interactions and uses deep learning to work their way through larger amounts of chemical space instead of existing literature.

Bringing Together Science and Technology

Matthew Might
Covid-19 has accelerated bringing together computer science with biologists. What has been missing is not more computer science, but more computer scientists. Hopefully one result of Covid-19 will be to bring more people into the field.

Peter Lee
The pandemic has forced everyone to face reality and have less irrational exuberance. The collaborations and learning taking place right now have been faster and more useful than in the past.

Final Thoughts

Karen Akinsanya
Bringing together science and technology in a more meaningful way has the chance to really change the industry though the coming together of interdisciplinary teams to accelerate work being done..

About the AI in Biopharma Digital Series

Connecting biopharma’s science and technology is what we do at AI in Biopharma. To continue this important discussion, with a focus on Biopharma Data Management Strategies to Effectively Use AI Tools, register for the upcoming interactive digital summit on September 15th. Speakers include thought leaders from: AbbVie, Novartis, Regeneron Pharmaceuticals, Inc., Unlearn.AI, The APANDEMIC Initiative, and moderator Harry Glorikian, General Partner, New Ventures Funds.

Contact the AI in Biopharma Co-Founders with questions or suggestions
Debra Chipman: [email protected]
Elizabeth Cutler: [email protected]

If you are interested in sponsorship information for an upcoming AI in Biopharma digital event, I would love to hear from you. You can contact me directly at: [email protected].

/* */