The dataset the field has been missing.
Microbiome science runs on snapshots: one sample, one moment, one study that ends. Source Biome is building the opposite. An open, anonymized dataset that follows real people over years, that any researcher can actually use, and that gives back to the people who make it possible.
A loop that pays everyone back.
You contribute
Every test and journal entry you keep in Atlas can join the commons in anonymized, aggregated form. Your data stops being a dead end and becomes part of something larger.
Researchers study it
Instead of scraping together tiny, one-off cohorts, scientists get a large, longitudinal dataset on transparent terms. That means better-powered, more reproducible research.
It flows back to you
The findings return to your account as sharper context: what the research says about your bacteria, how your changes compare, what the community looks like. You benefit from the data you helped build.
Time is the missing dimension.
Almost every microbiome dataset is a snapshot: a single sample, frozen in time. But the microbiome is a moving system. It shifts with diet, seasons, illness, medication, and age.
Longitudinal data, the same people measured again and again over years, is what lets researchers separate signal from noise and cause from coincidence. It is also the hardest kind of data to collect, which is exactly why so little of it exists. The commons is built to change that, one member at a time.
- ·Longitudinal, not a one-time snapshot
- ·Vendor-agnostic and normalized
- ·Method-aware (16S vs shotgun kept separate)
- ·Open and FAIR, not locked in a company
- ·Governed by a public protocol
It grows as we do.
Aggregate gut composition
To protect privacy, aggregate figures appear only once at least 30 members contribute, and are always shown with confidence intervals, never as bare estimates. We are still building toward that. Create an account and keep contribution on to help the commons reach this threshold sooner.
- Contribution is your choice. You opt in when you create an account, and you can change what is shared or withdraw at any time.
- Only de-identified, aggregated data enters the commons. We never sell individual-level data.
- Aggregate figures appear only above a minimum sample size (N ≥ 30) and always with confidence intervals.
- Cohorts are pre-registered, so results are not fished from the data after the fact.
- The dataset follows FAIR principles under transparent, public governance.
For members
Create a free account, keep your data in Atlas, and your contribution helps the commons grow while sharpening the context in your own account.
Create your free accountFor researchers
Apply for access to the commons. Tell us about your work and how you would use the data, and we will follow up.
Request data access