TL;DR: A major bottleneck in biomedical research involves tedious routine work performed by highly educated researchers. Lab automation lags far behind other industries. UniteLabs offers plug-and-play solutions. Their technology is deployed at Idorsia. The SiLA standard enables universal device communication. Current incentive systems discourage the collaboration needed to accelerate progress.
About Oskari Vinko
Oskari Vinko is cofounder of UniteLabs, a Basel-based lab automation and robotics company. His background spans mathematics, engineering, biotech, and synthetic biology. Follow Oskari on Twitter or connect on LinkedIn.
How UniteLabs Started
The idea for UniteLabs was born at ETH Zurich, where Oskari was doing synthetic biology research. He noticed that highly trained PhD researchers were spending enormous amounts of time on tedious manual tasks — pipetting, sample preparation, data entry — work that in other industries would have been automated decades ago.
Current lab automation is rather primitive compared to mobile phones and self-driving cars. We are asking people with PhDs to spend their days doing repetitive manual tasks that a robot could do better and faster.
The Customer: Lab Workers Without Robotics Training
A key insight behind UniteLabs is that the people who need lab automation most — bench scientists in pharma and academia — typically lack training in robotics and computer science. The platform is designed to be plug-and-play, requiring no programming expertise to set up and operate automated workflows.
On-Site vs. Cloud Labs
When asked about cloud lab services like Transcriptic (now Strateos), which allow researchers to run experiments remotely in centralized facilities, Oskari explains why on-site automation remains preferable for many use cases:
- Transportation of delicate materials: Many biological samples cannot be easily shipped without degradation
- Privacy of intellectual property: Pharma companies are reluctant to send proprietary compounds to external facilities
- Real-time control: Researchers need to monitor and adjust experiments as they run, which is difficult with remote labs
Case Study: Idorsia
Idorsia, a Swiss pharmaceutical company, has deployed UniteLabs technology to automate compound analytics. The system handles high-throughput screening, liquid chromatography, and mass spectrometry workflows — processes that previously required significant manual intervention and were prone to human error.
Device Communication: SiLA 2.0 Standard
One of the biggest obstacles to lab automation is that instruments from different manufacturers cannot easily communicate with each other. The SiLA 2.0 standard addresses this by providing a universal communication protocol for laboratory devices — essentially acting like Bluetooth for lab equipment.
Key features of SiLA 2.0:
- A universal protocol that enables any SiLA-compliant device to communicate with any other
- A universal experiment description language that standardizes how workflows are defined
- Integration with platforms like protocols.io for sharing automated protocols
- Vendor-neutral design that prevents lock-in to any single manufacturer's ecosystem
Other Bottlenecks in Biomedical Research
Beyond lab automation, Oskari identifies several additional bottlenecks that slow down drug discovery:
- Negative results are not shared: When experiments fail, the results are typically filed away rather than published, leading other researchers to waste time repeating the same failed approaches
- Routine work requiring human judgment: Some tasks, like protein crystallization, involve tedious repetitive work but also require expert judgment calls that are difficult to fully automate
- Lack of cross-disciplinary collaboration: Scientists tend to stay in narrow disciplinary boxes, missing opportunities for breakthroughs that come from combining perspectives
The War for Reputation: Why Collaboration Fails
When asked what prevents scientists from collaborating more, Oskari points to the academic incentive system as the root cause:
A professor requires four first-author papers to get tenure. This 100% disincentivizes collaboration. Why would you share your data or help someone else's project when your career depends on being the sole author?
This "war for reputation" creates a zero-sum game where sharing and collaborating are punished rather than rewarded.
Better Models from Software Development
Oskari argues that the software industry has already solved many of the collaboration problems that plague academia. Two platforms stand out as models:
- Stack Overflow: Breaks contributions into smaller units. You do not need to write a full paper to earn credit — answering a single question earns you reputation. This lowers the barrier to contribution and enables micro-collaborations.
- GitHub: Enables post-publication collaboration through pull requests and issues. Code is shared openly, and contributions of any size are tracked and attributed.
These platforms share key principles that science could adopt:
- Smaller contribution units — not everything needs to be a full paper
- Post-publication collaboration — building on shared work rather than working in silos
- Peer assessment — community-driven evaluation of contribution quality
Example: Rachel Harding and Open Data Sharing
Oskari highlights Rachel Harding as an example of a researcher who generates real collaboration through radical openness. By sharing her data publicly and blogging about her research in real time, Rachel has attracted collaborators and accelerated her work on Huntington's disease — demonstrating that open science is not just an ideal but a practical strategy for better research.
About Bio2040
There are so many challenges in drug discovery. We are a group of entrepreneurs and scientists who want to improve things.
Subscribe to the Podcast on iTunes
Be in the know and sign up for the newsletter.
Follow Flavio on Twitter here.