VeriSIM Life (VSL), a San Francisco-based biotechnology startup, builds artificial intelligence (AI)-enabled biosimulation models to accelerate drug development. The company’s BIOiSIM tool is an in-house technology that enables simulation and prediction of drug disposition in human and animal bodies.
The company’s goal is lofty. VSL is attempting to model whole human biology to solve, in part, drug development shortcomings that exist in animal testing by developing what’s called in silico simulations, as distinguished from in vivo and in vitro procedures. Animals models are frequently poor representations of human response. Plus, there is the whole cruelty issue when animals are exposed to all the bad testing outcomes common to drug discovery.
According to Yogesh Naren Bundey, a machine learning engineer at VSL, models trained on limited datasets can have skewed learning, overfitting and low optimization problems which directly affect their accuracy and generalizability. To put it in simple terms, because of the limited historic data available for AI to learn, models cannot properly distinguish the nuances from diverse populations and suffers undue significance to specious correlations. Therefore, small and sparse datasets have become one of the major blocks to AI gaining traction in pharmaceutical development.
Improvements in the prediction accuracy of BIOiSIM reduce costly drug failures at earlier phases of development. This is vital to VSL’s mission to drastically reduce drug attrition for pharmaceutical companies as 92% of drugs that reach human trials are not successful. The global drug discovery market is estimated to be worth at least $35 billion.
VSL recently raised a $5.2 million series seed co-led by OCA Ventures and Serra Ventures with participation from Susa Ventures, Intel Capital, Stage Venture Partners, Village Global, Twin, and Loup Ventures.
With this new funding, VSL plans to optimize its models through expanded academic partnerships, grow a team in the engineering and operations space, and engage with greater agency when working with larger pharma institutions.
VeriSIM was founded in 2017 by Dr. Jo Varshney, a multi-disciplinary leader in veterinary & human health development. Together with a team of world-class scientists, machine learning engineers and in silico simulation experts the team is bringing clarity to drug development through computational sciences rather than the existing trial and error experimentation approach.
With a machine learning driven approach, VSL is able to “pre-flight” advancements in medicine without the need for redundant trials, subjecting animals to the cruelty of testing, and achieve results with far better outcomes than the 8% success currently seen in medicines making it to market.
Here is an informative podcast interview with Bryan Vicknair and Jason Walsh of VSL.