Recursion is a clinical-stage biotechnology company decoding biology by integrating technological innovations across biology, chemistry, automation, machine learning and engineering to industrialize drug discovery.
One of the aspects of our technical culture that we’re most proud of is fostering transparency in the data work that folks are doing. This unlocks serendipity and opportunities to collaborate with colleagues that are working on questions or issues we’re interested in tackling. Hex’s Knowledge Library has really been transformational in this regard.
Many times, I’ve been thinking about a particular problem and, while scrolling through the Library, discovered relevant projects that others were working on. I’ll then either pull inspiration from the existing work or reach out to the authors to see how I can help. Once I’m engaged with the project, it’s easy for me to collaborate with others, whether I’m just leaving feedback in comments or adding code. Even if I don’t engage with the project, knowing that the project exists will prevent me from duplicating previous work.
With our company’s hybrid working model, this sort of transparency and asynchronous collaboration becomes all the more important. We have a mixture of folks who are entirely online and who are in offices brushing elbows with one another on a regular basis. Hex gives us some sort of common ground. It’s kind of like a watering hole where everybody can go and explore what other people are up to. It makes it easy to see what others are working on, have worked on, and could use help with.
Another area where Hex has been instrumental for us is in bridging the communication gap between biologists and data scientists. Recursion’s mission requires a lot of back and forth between these two groups. Before Hex, a biologist would ask a data scientist to answer specific questions using data, the data scientist would respond with a static report, the biologist would request changes, the data scientist would produce a new static report, and so on.
Hex has acted as a collaboration accelerator, enabling a much more streamlined interaction between data scientists and biologists. Biologists can provide pointed feedback with Hex by adding comments to a published Hex app. Then, I can go into a project, see precisely where the question is coming from, and often implement a slight change in minutes instead of generating a new report. This workflow has dramatically improved communication and efficiency between these two seemingly disparate teams.
Our biologists love using Hex because it’s so simple to create interactive apps. Before Hex, data scientists would tag-team with biologists to create several static versions of a particular analysis in separate notebooks. They would send these notebooks over to a data engineer, who would build an interactive app using a Python framework for the biologists to use. This entire process was time-consuming, and the friction prevented more interactive applications from being built in the first place.
With data apps, there’s virtually no friction for our data scientists to create interactive apps in parallel with the analysis they’re doing in Hex. It’s as easy as creating input cells and dragging-and-dropping.
Hex seamlessly fits in with our collaborative culture; it helps bring data scientists and biologists together, which is essential for the kind of data-centric drug discovery work that we’re proud of. Hex supercharges our ability to bring medicines to patients faster.
Genevieve Roberts, Senior Data Scientist