As we are perhaps reaching the end of a cycle of innovation in tech – the one that resulted from the simultaneous emergence of social, mobile and cloud – and collectively pondering what’s next, one of the areas I’ve found particularly exciting recently is the intersection of Big Data and life sciences.
A little over two years ago, in connection with my investment in Recombine, a genomics startup, I wrote (here) about another powerful combination of trends: the sharp drop in the cost of sequencing the human genome, the maturation of Big Data technologies, and the increasing commoditization of wet lab work.
The fundamental premise was, and still very much is, as follows:
- Cheap genome sequencing will unleash massive amounts of data (about a hundred gigabytes of data per human genome!);
- Big Data technologies will make it possible to process those massive amounts of data quickly and cheaply enough;
- Machine learning and artificial intelligence will enable deep analysis of the data;
- The hope is that through this analysis we’ll discover new insights in human health, and will be able to provide personalized treatment that will focus on every patient’s specific genetic makeup and needs.
Clearly, this premise is not new, and genomics has been around for decades. It has already seen several generations of startups, and its fair share of ups and downs, hopes and broken promises — we’re currently on an “up”, with strong signs that the genomic era is now arriving, and that this time it’s probably real.
But, at least from a software/internet VC perspective, one aspect is reasonably new: increasingly, the new genomics companies (and a number of other bio startups beyond genomics) look like software companies, rather than life sciences companies. They think about the world in terms of software and data. They have cross-disciplinary teams that include a lot of software engineers. They iterate quicker. They’re fundamentally data-driven. They require much less capital to start. They can increasingly leverage cloud resources, as both giant tech companies (Amazon, Google) and startups (DNA Nexus) are beefing up their offerings to store and analyze genomics data in the cloud.
To put this in context, I want to talk about two companies I’m very familiar with, as an investor: Recombine and Phosphorus.
Recombine was co-founded by Alex Bisignano, an impressive young CEO with a broad background that includes biology and computer science, and Santiago Munne, a widely-published scientist in the field of reproductive genetics, and a seasoned entrepreneur.
Recombine is a great illustration of how capital-efficient this new generation of companies can be. Recombine raised a small Series A of $3.3M in early 2014 (from FirstMark, with institutionals like Vast Ventures and angels like Nat Turner, Zach Weinberg, Vivek Garipalli and Alexander Saint-Amand participating). It quickly built a great team of 60+ engineers, research scientists, genetic counselors, operations, sales and marketing people. Recombine did not own a wet lab, choosing instead to partner with a third party. As the company’s annual revenues rapidly ramped up to 9 figures, for a long time it didn’t touch much of the money we had invested. A few weeks ago, barely two years after their Series A, Recombine’s business was acquired for $85M by The Cooper Companies.
After this great outcome, Alex and Santi decided to immediately double down and go back at it, with some of the team and technology that were not part of the Cooper acquisition.
Their new computational genomics company is called Phosphorus, and it’s even more ambitious than Recombine. The company is launching today, and I’m excited to be leading a $10M Series A investment in it.
Phosphorus is very emblematic of what this new generation of genomics companies looks like:
- Over 80% of its early team is comprised of computer scientists, data scientists and research scientists, including a number of PhDs;
- It leverages a fully modern Big Data architecture (Spark, etc.)
- It’s building software that will enable labs to run their own genetic tests
- Because any data company needs a “data trap” to quickly build a meaningful data set, it’s also seeding the market with new genetic tests, starting with reproductive medicine but with plans to expand quickly to other areas
- It is working with third party labs, as opposed to building its own lab
- Ultimately, both the network of labs using the software and the genetic tests it offers contribute massive amount of data to Phosphorus
- Phosphorus can use the data to run data science at scale, build new products faster, and make participation in the network more attractive to new members, essentially creating a data network effect.
Of course, it’s important to not exaggerate the trend – while they increasingly look like software companies, these new genomics companies (and other bio startups) are still part of the life sciences world. As such, they will have almost inevitably to deal, directly or indirectly, with regulation and the various intricacies of the US healthcare system – IRBs, CLIA, FDA and payor network relations are still very much part of the conversation.
But overall, I believe we are in the early days of this new wave of innovation in the field – as the cost of sequencing keeps dropping, and the power of Big Data and AI keeps increasing, we’ll see more and more companies with similar characteristics. We’re not quite at the stage where a couple of biology majors could start a genomics startup from a Starbucks, and perhaps we’ll never get there, but the trend is nonetheless exciting.