Caffeine Wrap Image Caffeine Wrap Image

Harnessing biological insights from 50,000 peer-reviewed papers to make sense of pre-clinical, patient, or consumer product-driven data


Selventa uses a Forensic Approach to deliver deep molecular biological insights into your data, by comparing it against results from 50,000 published experiments

  • A comprehensive knowledgebase of cause and effect relationships
  • Analytics that expertly query the knowledgebase to provide valuable information about your data
  • A visualization platform using Cytoscape, where results can be manipulated using additional tools

Selventa Analytics

Listed are examples of the different ways our analytics help us understand your data. The links below each point to examples where that particular analytic was either developed or employed


We’ve worked extensively with RNA and DNA sequencing data, together with clinical measurements, proteomics, phosphoproteomics, metabolomics, etc.

Our platform allows us to rigorously evaluate pre-clinical models for suitable translatability to human disease to de-risk failures in patients. We can identify, on a molecular basis, the strengths and weaknesses of each model. Our ethos is to let patient disease biology drive pre-clinical model testing, not the other way around.

We are experts at maximizing the value of public patient data streams, if you have not collected your ideal data yet, and there is suitable public patient data to answer your question, we will incorporate it into our analysis. If you have data from pre-clinical models, we can use that as well.

While every solution we offer is uniquely tailored to the question we are answering, we can take three different approaches depending on what data is available.

  • Using solely our knowledgebase and expert traversal algorithms, requiring no external data

  • Using algorithms designed to identify the mechanisms (potential targets) that would have to be modulated to return a disease state to a healthy state

  • Using a data-driven approach capable of identifying novel biological connections that have not yet been described in the literature

The challenge here is often that limited molecular and response data is available to do robust statistics for biomarker discovery and to inform further clinical design through appropriate patient selection. Selventa harnesses all relevant available data (e.g., from pre-clinical toxicology studies, IHC, public disease-relevant patient data, plus any clinical trial data) to do preliminary biomarker discovery.

This is an excellent time not only to begin selecting patients most likely to respond, but also to understand why patients are not responding, and generate hypotheses about how to sensitize these patients to treatment that can then be tested in pre-clinical models.

We use the same principles as we would for earlier phase biomarker efforts, but often have access to more patient samples, and asset-relevant response data. Our approach is to not only provide the content for the assay, but also to provide the molecular context for each measurement in the assay. For example, linking it biologically to the disease and/or the MOA of the asset.