We are excited to share more about why we led A-Alpha Bio’s Series A round announced today and to congratulate David, Randolph and their team on their vision and progress to date. A-Alpha is a platform life sciences company unlocking the opportunities to understand and leverage Protein-Protein Interactions (PPIs) for drug discovery. They are building a system, combining a high-throughput synthetic biology platform with machine learning, that dramatically improves the industry’s ability to engineer antibodies and other life-saving drugs. They are already delivering the most scalable and sensitive measurement of PPI’s available today!
Throughout the last five years, Madrona has been exploring the convergence of life and computer sciences and are excited about the opportunities being built at the “Intersections of Innovation” in these areas. We are also accelerating our investment activity following earlier investments in Nautilus Biosciences, Twinstrand, Ovation, and others. We believe that companies utilizing novel approaches for exploring these intersections see the future differently and are poised to create the next set of breakthroughs in the life sciences. This area of increasing investment focus and experience for Madrona is also poised to have a significant, positive impact on our lives and society. As machine learning, computing power, and wet lab automation advances converge, we expect to see even more companies approaching research problems in paradigm shifting ways. Our focus is on companies that are either creating ML-optimized novel data sets resulting from their lab techniques and/or applying ML to existing datasets in new ways.
We believe that companies utilizing novel approaches for exploring these intersections see the future differently and are poised to create the next set of breakthroughs in the life sciences.
A-Alpha Bio is a platform life sciences company, founded by David Younger and Randolph Lopez, focused on screening and measuring large networks of Protein-Protein Interactions (PPIs). For example, Antibody/antigen interactions are a familiar type of PPI that researchers are extremely interested in understanding in more detail to treat diseases like Covid-19 and cancer. While there is existing high-quality data on protein sequences and structure, binding interactions between proteins have been difficult to measure at scale and traditional methods have proven slow, inefficient and imprecise. Furthermore, PPIs are extremely important in the function of cells and biological systems so mapping and characterizing these dynamics is a key challenge for life science researchers and pharmaceutical companies.
Historically, scientists were limited to methods that only measure binary interactions: screen two proteins in a single experiment and attempt to measure whether they bind to each other. While these methods have led to a deeper understanding of PPIs – especially if scientists have an idea of what to look for – they come with drawbacks: data fidelity can be low, looking at single pairs of proteins in isolation eliminates important variables like competitive binding dynamics, and the process has slow cycle times. The founders of A-Alpha Bio developed the AlphaSeq platform during their graduate research with David Baker and Eric Klavins at the Institute for Protein Design and Center for Synthetic Biology at the University of Washington to solve these problems and unlock the ability to screen and analyze millions of PPIs simultaneously and in the same experimental environment.
Unlike binary methods, A-Alpha’s process results in an unprecedented amount of data – each run of 1000×1000 protein libraries generates 1M PPI data points that can be ingested into the A-Alpha ML model and used to iterate and improve the next run. Each iteration helps create a “flywheel effect” by adding to a unique dataset which can then be trained to understand and predict PPIs in greater detail. The method also allows A-Alpha to upend aspects of the traditional drug discovery process – rather than focusing on a single target and iterating slowly on potential antibodies, A-Alpha can screen large libraries of antibodies against all variations of a target in parallel to create truly differentiated insights and data. Their work on identifying the best broad binders across the set of Covid-19 variants is a good example of this unique capability. A-Alpha’s technology is exciting and well positioned to generate novel insights across a wide range of problems.
We invested in A-Alpha’s seed round in 2019 and have been consistently impressed by the progress the team has made on both the scientific and commercial fronts. From initial incubation at the Institute for Protein Design to the advances made over the last several years, David and Randolph have proven the power of their approach and are poised to accelerate quickly. And finally, they have assembled a truly special team that is building an industry changing company. We could not be more excited to be deepening our partnership with them as their investor today and for the exciting journey that lies ahead.