NO-CODE
The code lives behind the node
You work with a canvas and parameters, not conda environments and job scripts. The person who generated the data can finally run the analysis.
Early access · onboarding labs
Build bioinformatics pipelines, visually.
OpenBionode is a no-code tool for building and running bioinformatics pipelines. Connect the steps scientists already use, and a deterministic engine runs them, so the same inputs give the same outputs on any machine, a year later.
Like n8n or Zapier for pipelines, but deterministic.
Works with the formats you already have:.fastq.bam.vcf.mzML.counts
rnaseq.flowrunning
ResultVolcano plotsame hash on re-run
The problem
Bioinformatics is bottlenecked by tooling, not biology.
The data’s owner can’t run it
Biologists generate the data. Analyzing it means conda, cluster scripts and a Nextflow file, so it becomes someone else’s job and they lose control of their own experiment.
Generated code you can’t read
AI will write a pipeline in a minute. But if you can’t read code, you can’t tell whether it’s right, change one parameter, or get the same numbers twice.
A log is not a re-run
Most tools treat provenance as logging. A log tells you what happened. It doesn’t get your numbers back. Re-executing an analysis is a different problem.
The whole analysis, on one canvas.
Everything a run needs in one place: the steps, the parameters, the compute, and a record precise enough to reproduce any result.
DETERMINISTIC
Same inputs, same result
Every run records the exact tool versions, inputs and environment. Re-run it on another machine or a year later and get the same numbers.
TOOLS
~3,000 steps, ready to connect
The aligners, callers and quantifiers scientists already use, ready to drop onto the canvas. Connect them there instead of wiring them together in Bash.
RUNTIME
Local now, cloud next
Runs on your laptop today as a desktop app. The same pipeline scales to the cloud when your data outgrows one machine.
AI, without the risk
AI can assemble a pipeline, or write a step that doesn’t exist yet. It never runs one. A deterministic engine does, so your results never depend on which model wrote them.
From data to a result you can reproduce.
- 01Step
Build on the canvas
Drag in the steps you need and connect them. Set parameters in a panel, not a config file. An AI assistant can draft the pipeline for you.
- 02Step
Run it
A deterministic engine executes each step and records the tool versions, inputs and environment as it goes. Local now, cloud soon.
- 03Step
Re-run it, anywhere
Open the same pipeline on another machine or a year later and get the same result, down to the byte.
Bring the analysis you can’t hand to anyone else.
OpenBionode is in early access. Join the waitlist, tell us what you analyze, and we’ll prioritize the steps you need.