Digital biotech company OccamzRazor recently raised $6.1MM in a series A funding round that closed in March 2022. The round was led by Jeff Dean, the head of Google AI, a division of Google dedicated to tackling healthcare problems and advancing scientific discovery by leveraging computer science.
OccamzRazor is a biotech startup on a mission to cure brain aging through digital science. The company uses natural language processing, graph prediction, and graph machine learning to map out disease processes related to aging, such as Parkinson’s Disease. The company claims that human capacity is not sufficient to translate all the public knowledge into clinical trial success. Their method is to use AI to analyze data and test out novel treatment targets at a faster rate and lower cost than traditional pharmaceutical and biotech companies. Specifically, the algorithm attempts to make links between data extracted from resources like PubMed and other databases to create non-obvious links between diseases and drug targets. The company refers to all their compiled information as “Parkinsome,” a 3D graph that details all proteins, cells, metabolites, and genetic components related to Parkinson’s Disease. CEO and co-founder Katharina Sophia Volz, PhD, says that “We are currently in the middle of an exponential increase in biomedical datasets that can offer new perspectives on neurodegeneration. Hidden within this data are the insights needed to develop the right therapies for the right patient populations — the data just needs to be connected in the right way, with the right AI built to interpret it.” OccamzRazor is currently at the pre-clinical stage of their first AI-generated therapies.
The company was founded in 2015 by German medical researcher Katharina Sophia Volz. Volz had recently completed her education at Stanford University, where she was the first PhD in stem cell biology and regenerative medicine. Her call to action came one day when she received a message that someone close to her had been diagnosed with Parkinson’s Disease. That’s when she vowed to find a cure for the degenerative disease – starting with compiling all relevant data including clinical trials, publications, and genomics information. Volz collaborated with the Stanford Computer Linguistics department to develop the initial algorithm, which classifies biological parts and maps out how they interact with one another. The company’s name, OccamzRazor, was coined from philosopher William of Ockham’s principle that states when solving a problem, we should gravitate towards the simplest explanation with the fewest assumptions. Volz’ goal is to use this principle to simplify complex disease processes, starting with Parkinson’s Disease and eventually advancing to other neurological conditions. For her efforts, she was featured on the annual Forbes’ 30 under 30 list in 2017, one of many other accolades during her time as CEO. Volz’ team is rounded out by computer science engineers and other scientists including 2013 Nobel Laureate Randy Schekman. Schekman serves as an advisor along with multiple other leaders in the medical science field.
When asked why OccamzRazor’s initial efforts focus on Parkinson’s, Volz notes the disease’s significant unmet need. Parkinson’s patients have few treatment options- none of which are curative or have long lasting durability. From Merck to Roche, many big pharmaceutical players have attempted and failed to achieve late-stage clinical trial success for Parkinson’s therapies. Many companies have abandoned efforts to tackle the disease all together, making way for smaller startups to crop up. OccamzRazor is among many other agile, digital companies striving to tackle neurodegenerative disease research with machine learning and artificial intelligence. Although the possibilities of these applications are exciting, the technology is still in its infancy. One of the largest setbacks for machine learning in the neurodegenerative space is insufficient datasets. Machines rely on their inputs, collated data, in order to produce accurate, consistent, and generalizable results. Patient metadata is particularly susceptible to human error, participant drop out, and inconsistent measurements. This makes careful data curation an essential step when preparing machine learning algorithms. In addition, knowledge graphing, which makes associations between different concepts, can sometimes lack granularity. Programs must have high specificity in order to be useful in the setting of disease pathophysiology. OccamzRazor claims to overcome this obstacle through its unique biomedical information extraction (IE) pipeline. They maintain that their IE can augment novel disease-gene association predictions by a relative 20%, even when using a small amount of machine training data. This differentiation supposedly allows the company to increase the program’s predictive accuracy and ultimately unveil hidden biological relationships.
Since its founding, OccamzRazor has raised over $12MM in funding from investors like Lauder Partners, Valor Equity Partners, Bioverge, and re.Mind Capital. Supporters and collaborators include The Michael J. Fox Foundation, The Neuroscience Research Institute, and Parkinson’s Institute and Clinical Center. Regarding the latest funding round, Investor Jeff Dean notes “I’m excited to see the progress OccamzRazor has made thus far on integrating many disparate sources of data and applying machine learning towards the discovery of new candidates for fighting degenerative brain diseases.”