AI-drug development platform BioPhy launches with $4.5M

The company's operating system accelerates the ID and development of promising drug candidates

When it comes to evaluating the validity of science, it's still a really opaque and qualitative thing to do: you can have subjective assessments of whether or not something is going to work in the clinic, or whether or not the particular formula for a drug is what's needed to achieve a desired result, but that's very limited.

That's what led Dave Latshaw II, PhD to co-found BioPhy, which was formed the thesis that we needed to come up with a better way to make these evaluations. The company, which uses AI for drug development, officially launched on Tuesday, along with $4.5 million in funding from Metodora Ventures and Audere Capital, as well as "prominent figures in life sciences" including Jeff Marrazzo, a co-founder of Spark Therapeutics.

"The human and individual person can only spend so much time researching those areas, which means that for very hard problems like this, you need a lot of people, a lot of time, and a lot of resources to solve that problem," he said in an interview.

"In the end, you still only have a qualitative answer. So, what we wanted to do was create a platform that could identify the most promising science and then, once you have done that, accelerate the process of developing that, so it can get to patients."

The company’s AI operating system for drug development currently consists of two products. the first is BioPhyRx, a generative AI solution designed to create a centralized, intuitive research environment for accessing scientific and regulatory resources. It uses large language models to help pharmaceutical companies by analyzing and interpreting scientific literature, clinical trials, regulatory guidelines and submissions, and other industry specific sources to provide accurate and up-to-date information on demand. 

The second product is BioLogicAI, a predictive AI engine that provides customized insights to aid life science companies through the drug development process, including clinical trial endpoint predictability, indication selection, licensing, drug repurposing, asset acquisitions, and divestment.

For example, a pharma company that has developed a new drug will want to differentiate themselves from competitors. The BioPhy platform is able to put them in a very transparent light relative to those other other drugs by quantifying the efficacy, quantifying the pathways that the drugs act through, and their feasibility in clinical trials and in people.

"Where we slot in is once the drug candidate has been identified. Before that, there's a lot of work that goes into designing and optimizing the compounds to get them to the place where you want to begin more rigorous forms of testing," Latshaw explained.

"This validation is usually done in the lab. Over the years, more robotics have been used to do screening and things of that nature but what we really bring to the table is being able to do computational identification and benchmarking of assets."

BioPhy works with pharma companies at various stages of their drug development; if a company is at an earlier stage then BioPhy is typically focusing only on validating the biological thesis through a multi-omics system it developed as a part of its biologic AI platform. That includes a quantitative scale where it calibrates the scores of drugs that have been approved, or not approved, by the FDA, including the interaction between proteins and different molecular pathways. Then, when it does predictions for the new drug, it can see which side they skew towards, and how much of a gap they close relative to the things that have already been approved or what's already in development.

"Because we can do this across all other drugs, we're able to show where their drugs stack up against drugs that are either in development or already approved by the FDA. And that gives them the ability to orient themselves to the competitive landscape and differentiate themselves," said Latshaw.

For drugs that are further on in their development, meaning they're either going into or are already in clinical trials, BioPhy brings in more operational data, including around the underlying patient populations for the trial, how the trial is structured, where it is, the capabilities of the people that are running the trial, and the management teams.

Together, those two products form the basis for determining how likely individual drugs are to pass their clinical trials, which helps pharma companies save money by showing them where to allocate their resources. If they put the money behind programs that are expected to be the most impactful or have the highest probabilities of success, that will leadd to much less wasted capital. 

"It takes a lot of money to bring these drugs to market and all that can be repurposed to bring the most promising candidates to patients," Latshaw said. 

BioPhy currently works with two of the top five pharmaceutical companies in the world; its platform has been making predictions on clinical trials that have been currently ongoing for the last two years and it has been able to successfully predict over 1,300 clinical trials at an 80% accuracy rate. Clients utilizing BioPhy’s patented technology have experienced up to a 100x increase in the speed and accuracy of scientific and clinical research, quality assurance, and regulatory intelligence in drug development. Its platforms also deliver a 5x enhancement in predicting and guiding clinical trial success across all endpoints and phases.

Now that it has raised this funding, the plan is to use the funding to build out its team, growing from 10 employees to 20 over the next year.

"It's probably no surprise that our company is very tech heavy, so the idea is to put a lot of the capital towards scaling our engineering teams, especially because we're trying to meet some growing enterprise demand. And, beyond that, building out the team to engage with new potential partners," said Latshaw.

In the longer term, the company is already looking at potential applications outside of life sciences, and is working U.S. government and intelligence agencies, financial services, and the public sector. For example, on the government side of things, one of the organizations that BioPhy works with is interested in trying to identify the most promising companies to attract from an economic development perspective. In financial services, meanwhile, the company can help them solve their capital allocation problems, which is something its also done for life sciences companies. 

For now, though, the company is putting its the majority of its focus on life sciences, with the goal of being so vital to the industry that no one wants to make a decision without having its metrics. The reason it can do that, said Lathsaw, is because its AI is currently fully operational.

"A lot of times when people think of artificial intelligence, they think of some abstract concept that can do really interesting things, but is unable to create concrete value, maybe a parlor trick, so to speak. We are certainly not that," he explained.

"I'm very grounded in practicality being an engineer by background, so we like to think of this as practical AI. What I mean by that is you can create value with it directly today: it’s not a hypothetical, it's not something that's going to take five to 10 years to materialize and create value because nobody really knows what to do with it. It's something that has direct applicability, it's something that's been validated for years at this point, and it's usable immediately by our partners."

(Image source: biophy.ai)