Pharmaceutical development and manufacturing are increasingly reliant on real-world data, analytics, and modular AI architectures. An example of the application is with drug–patient compatibility, reduced trial attrition, and to help to revive stalled or shelved assets.
BullFrog AI CEO Vin Singh’s company has developed a proprietary platform called bfLEAP™. The platform uses explainable, graph-based AI to analyse complex, multimodal datasets and uncover patterns that traditional statistical methods often miss. Digital Journal caught up with Singh to learn more.
Singh is a serial entrepreneur with extensive experience in the healthcare industry. Previously, he was the Founder & former CEO at Next Healthcare Inc., Co-founder of MaxCyte Inc. (Nasdaq: MXCT), a company with a $1.5 billion market cap, and Global director of cell therapy at ThermoFisher Scientific.
Digital Journal: Can you tell us a little bit about BullFrog AI and your technology solutions?
Vin Singh: BullFrog AI was created with the idea to use AI and machine learning to overcome a costly problem in the drug development ecosystem: stubbornly low success rates in drug development, despite advances in biology and data generation. Too often in our business, promising therapies fail, often late in their development, because earlier decisions were required with incomplete or missing information. From the start, our vision has been to help life sciences companies make better decisions earlier, reduce wasted capital, and ultimately bring more effective therapies to patients faster.
There are a lot of companies talking about AI in drug development, but none have our core platform from Johns Hopkins Applied Physics Lab, and very few, if any, have our causal inference capabilities. Our bfLEAP® platform is designed to uncover critical biomarkers, disease drivers and pathways, not just patterns. That distinction matters because it leads to insights that are more actionable and trustworthy.
DJ: What are your key AI products and what differentiates them from others in the market?
Singh: Our end-to-end solutions in this space are bfLEAP®, bfPREP™, and BullFrog Data Networks®. bfPREP™ prepares large-scale datasets for AI consumption, a service that a lot of these companies need. This then feeds directly into our core platform, bfLEAP®, which is a causal AI system designed to analyze complex biological and clinical data sets. What makes it different is its ability to work with incomplete, shallow and wide, and multi-modal data, which is the reality of drug development data.
bfLEAP® then feeds into BullFrog Data Networks®, building upon the AI analytics engine to uncover patient subgroups with shared molecular signature, high-impact genes and pathways driving disease biology, and much more. This end-to-end approach, from preparing the data all the way to providing key insights that can enable better-informed decision-making at key moments in the drug development process, is ultimately what separates us from others in the market.
DJ: What’s the difference between a general AI platform and a ‘causal’ AI platform like BullFrog’s?
Singh: Causal AI is different in that it can go beyond relationships and patterns to provide the magnitude and direction of the key correlations that it uncovers. The main difference between the two, especially in complex biological systems, is that causal AI can provide an understanding of the drivers of disease, as well as the associated pathways, which can be crucial to making discoveries that will lead to successful drugs discovered and developed in less time and for less investment. At BullFrog, we are determined to provide more insight to researchers beyond just identifying correlations, relationships, or patterns.
DJ: You recently announced data identifying a 3x increase in survivability in pancreatic cancer data, can you talk about how BullFrog was able to achieve that and what that means for patients?
Singh: Absolutely, we recently completed an analysis for Eleison Pharmaceuticals and their phase 3 pancreatic cancer study. We ended up analyzing an incredibly large, complex, and messy data set, which resulted in the identification of biomarkers and a patient subgroup that led to an almost three times improvement in overall survival. These results were submitted and accepted for presentation at the American Society of Clinical Oncology’s Gastrointestinal Cancers Symposium (ASCO-GI) meeting in January of this year.
Eleison and their partners were very happy, and they will decide what is best to do with the findings from that study, but their conclusions for the presentation were that our platform is capable of successfully identifying patient subgroups within existing clinical trial data. We hope those insights can make a big difference for patients suffering from such a horrible disease and can be applied elsewhere.
DJ: What’s next for AI-integration in the drug discovery process?
Singh: We recently announced our new capability that will be launching on March 25, which will slot into our end-to-end solution with a scenario-based design engine that can clearly define clinical trial strategies and build diversified, risk-balanced R&D portfolios. We’re very excited to see how some of our existing and new clients will be able to apply this new solution to enable robust recommendations and highlight clear pathways for strategies with a higher probability of success.
More broadly, AI integration is only going to become more accepted in this field, but it’s important we’re doing so with patients in mind. We can’t just be applying AI for AI’s sake. At BullFrog, we really want to help educate the industry about how to properly prepare the right data to enable clinical trial teams to gain the right insights and further help with drug development success rates. The ultimate goal must remain improving and saving lives with better outcomes for patients.
