About RNA

    • 90% of the genome is transcribed to RNA: mostly to non-coding RNA such as micro RNA in humans (miRNA) or riboswitches in bacteria
    • Only 400 out of 22’000 human proteins are are proven to be druggable
    • Gene expression is regulated by miRNA
    • RNA has the potential for a huge new target space for drug discovery/Chemistry space unexplored/free IP space
    • Gene expression is regulated by riboswitches in bacteria
    • Proof of concept: Branaplam is in Phase II Trials for the disease SMA
About Saverna

Our Pipeline

As a lead development in our pipeline, Saverna will develop a small molecule to target pre-miRNAs for unmet medical need.


Our Drug Discovery Platform

Non-coding RNA, despite being reported to be dysregulated in over 1000 diseases, are notoriously difficult targets for drug discovery. Saverna Therapeutics’ drug discovery approach and process will impact the development of a new field in drug discovery by utilizing their innovative and advanced platform for identifying small molecule drugs targeting non-coding RNA, a novel class or drug targets. Saverna focuses on the disease areas of inflammation, autoimmune diseases, cancer and infection.

Our Approach


Saverna has integrated fragment-based screening (FBS) by Nuclear Magnetic Resonance (NMR), in-silico machine learning and cellular assays for drug discovery. FBS is a very efficient method to sample the chemical space with only a few thousand compounds. NMR data on fragments binding, or not binding to the target RNA will be used to generate models to identify larger compounds combining two fragments binding to proximate pockets of the RNA. These larger compounds will be purchased and analyzed in selectivity assays as well as in cellular and biochemical assays.

Our Workflow


Our trade secret proprietary workflow for RNA targets. It contains more than 50 machine learning models that have been trained with large amounts of data to predict the success of a hit-to-lead project and beyond. In addition, they are used to design and review libraries, annotation of compounds in collections, and to prioritize hits.

We focus on key problems that can
hurdle a drug discovery project.

  • Learn from new data
  • Learn from hits/non-hits
  • Learn from public data
  • Machine learning
  • Deep learning
  • ChemX
  • Induction rules
  • SAR cliffing
  • Seq2Frag
  • Probability to bind target
  • Probability to reach target in cell
  • Probability to reach off targets (tox)
  • Probability to pass phase ICT