Key points

  • Sapio ELN gives scientists a single interface for trusted analysis and design applications without replacing them. Sapio Elain, the AI co-scientist, coordinates the handoffs and returns results directly into the experiment record.
  • Scientists remain in control throughout. Sapio Elain handles the coordination; the scientist owns every decision.
  • The Elain ecosystem is not just an integration story. It is a governed, human-led environment where specialist science stays specialist and the notebook becomes part of the full design cycle.

Biopharmaceutical research teams are never short of computational and design software. Across bioinformatics, cheminformatics, genomic analysis and structure-based design, scientists already rely on trusted tools built for specific analysis and design tasks during the discovery process. These tools are powerful, validated by the computational experts within an organization, and often central to how research gets done.

The challenge is the environment in which they operate.

Every scientist understands the Design-Make-Test-Analyse cycle. Make and test happen inside the ELN where scientists live. Design and analyse have always required going somewhere else: running separate computational software or, more commonly, asking a specialist team to run it. Every handoff breaks the thread of the experiment, and every wait slows the next decision.

That is the real problem. It is not a lack of models or methods. It is a lack of a unified environment where computational and design tools can be applied directly to the research record by the scientist who needs the answer.

From passive records to active environments

Traditionally, the lab notebook was the place where work was documented after the fact. It needs to become the place where work is coordinated and analyzed while it is happening.

Sapio ELN is that environment. As Sapio Sciences’ AI Lab Notebook, it is not trying to replace the computational tools scientists already trust. Sapio Elain, an AI co-scientist that spans the entire Sapio Platform, works within Sapio ELN to bring those tools into a single interface so scientists can access them without leaving the experiment or waiting for a computational team to run it for them.

The scientist describes in natural language what they want to do. Sapio Elain coordinates the appropriate tool. The job runs through the trusted external application. The results return directly into the experiment record. The scientist reviews those outputs, decides what matters and determines the next step.

Sapio ELN is not built to prescribe a single best method for any design task. It is designed to expose the tools a scientific organization already trusts, validates, and governs so those methods can be used inside the environment where the scientist already works. The tools available through the Elain ecosystem are the tools each organisation has already selected, validated and invested in.

Sapio Elain is the intelligence layer that holds it together. Sapio ELN is where the experiment lives.

What this looks like in practice

At the recent Pistoia Alliance Annual Meeting, Rob Brown, Sapio VP and Head of the Scientific Office, demonstrated this using Reaxys from Elsevier. Starting from a set of candidate compounds already registered in Sapio ELN, he used Sapio Elain to calculate synthetic accessibility scores across the full compound table and return the results directly into the experiment record. A second prompt invoked retrosynthesis for a selected lead, returning proposed routes with confidence scores. Sapio Elain then created a medicinal chemistry experiment with links back to the underlying Reaxys records.

That is one example from one design cycle. The Elain ecosystem supports a broader chain of decisions across the full DMTA cycle. In practice a scientist would work through these questions iteratively, across multiple experiments over time; the example below compresses that into a single illustration.

A scientist might begin with shape-based search to identify promising candidates, move into early ADMET triage to filter on developability, apply rapid pose prediction or deeper docking to assess structural plausibility, and finally evaluate synthetic accessibility and retrosynthesis to confirm which compounds can realistically move into the lab. Each of those steps draws on a different partner integration and is covered in detail in the individual partner blogs in this series.

None of these steps are unusual on their own. What changes is that they no longer have to remain isolated from one another. The point is not that Sapio replaces solutions like OpenEye, Simulations Plus, NVIDIA, CCDC or Elsevier. The point is that Sapio Elain makes those specialist capabilities accessible inside a single environment, with results staying connected to the right entities, targets and experimental context at every stage.

The scientist stays in control

This is not automation that replaces scientific judgment. It is coordination that supports it.

All agent-coordinated outputs are explicitly flagged for review, with timestamps and provenance preserved in the experimental record. Sapio Elain handles the heavy lifting of invoking tools and returning results, but the scientist still owns every decision. They choose which question to ask, review what comes back, and determine what happens next.

Sapio ELN sits on the broader Sapio Platform, with a unified data layer and persistent identifiers for all entities, including large and small molecules and samples. That ensures results from one step remain tied to the right targets and experimental context in the next, making the research process more durable, traceable and reusable.

The Elain ecosystem

The Elain ecosystem is not just an integration story. It gives scientists a way to use the tools they already trust inside a governed, human-led environment, where results stay in context and each stage can more naturally inform the next.

In that model, the notebook is no longer just where research is recorded. It becomes part of the full design cycle.