To our shareholders,
There is a clear reason I decided to found Switzerland Omics. I kept asking the same question: when will we finally get the tools we actually need to make precision medicine reliable? For years, it seemed like the answer was that it was simply too hard. There were too many variables and too much uncertainty. But I knew it could be done. It just hadn’t been done yet.
Precision medicine, as it’s widely practised today, operates on the idea of “good enough.” Most tools focus on detecting what is present, typically the easy-to-find true positives, and then stop. But when it comes to life-changing decisions in clinical genetics, “good enough” is not good enough.
I believed that the most valuable contribution I could make was to take the harder path. That meant building a statistical framework that quantifies everything: what’s observed, what’s missing, what’s likely, and what isn’t. It meant starting from scratch, assembling foundational datasets, and refusing to compromise on depth or rigour. It also meant resisting the short-term incentives that often guide translational work. We chose not to chase one-off publication-ready results or focus on narrow disease subsets. We avoided dressing up partial insights as final answers.
Instead, we are building tools that will still matter ten years from now.
Switzerland Omics is a company built around evidence; capturing it, quantifying it, and making it usable in the real world. We develop statistical genomics methods that move beyond detection into interpretation. We replace assumptions with calibrated, transparent answers. Our models don’t just identify potential findings. They also assign a clear level of certainty to them, and just as importantly, to what isn’t found. We expand classical population genetics and Bayesian reasoning into a full diagnostic framework. For instance, in recent work, this means the result is not just variant-level reporting, but gene- and panel-level conclusions that are accountable and reproducible.
This year, we released a flagship dataset of calculated prior probabilities for every reliably-known disease-related variant across all major inheritance modes. To our knowledge, this had never been done at scale. We developed a method for combining observed and unobserved variants into a single probability of disease. We validated the model against national-level cohorts and showed strong alignment with real case data. We will continue to improve and validate results transparently. We also built new disease classifications based on protein networks and clinical phenotypes. These tools are now available to clinicians, researchers, and developers.
The core idea is simple. Precision medicine needs more than precision at the variant level. It needs confidence in the whole conclusion.
We are not optimising for flash. We are not chasing trends. We are doing the hard work to build the statistical infrastructure that will support the next decade of genomic medicine.
This is just the beginning. The foundation is set.
Dylan Lawless
Founder, Switzerland Omics
