Common questions¶
Find the next step for your slides, results and account.
Using Aurora¶
What does Aurora predict from a slide?¶
Aurora predicts spatial gene expression from routine H&E images, helping researchers explore molecular patterns across samples, cohorts and disease. Researchers can use the results to form hypotheses and decide what to validate experimentally.
Who can use Aurora?¶
Analysis is for academic and non-profit research. Commercial evaluation and use are governed by a written agreement with Aurora. Discuss commercial access.
What do I need before I start?¶
An H&E image of the slide.
To send a slide without writing code, upload it on the website with Aurora Direct.
To work from code, you need a machine that runs Python. Cutting and packing tiles needs no key. Embedding them on your machine, or submitting, needs one. Install and prepare a slide is the start.
Data and privacy¶
Do my slides leave my machine?¶
Only if you send them. Tiling, quality control and packing run on your machine, and so does reading a result. To keep every image there, send embeddings instead: Your slides stay with you is that route.
Does the package report anything about my work?¶
No. The package sends no telemetry: no usage counter, no crash report, no update check.
Where is my data stored?¶
Uploaded images, generated results and account information are stored on infrastructure in Switzerland, and analyses run there. Privacy and retention says how a request reaches it.
How long are my results kept?¶
A result file is kept for 90 days, then deleted. An upload you never complete is removed after a window the discovery response reports. Any other upload has no scheduled deletion.
Privacy and retention says what deleting a job removes.
Can Aurora use what I submit?¶
Under the academic terms, yes. Aurora may use the data you upload, and the results generated from it, to develop, train, validate and improve its models and services, including those it offers commercially. Your data is never made public, and deleting it does not withdraw a model already developed with it.
How Aurora may use it links the Terms.
Access¶
How do I get an API key?¶
Access is requested through the contact form on the website: Request programmatic access.
Say who you are and what you intend to do. Once access is granted, run
auroraomics login on the machine that will hold the key.
Get access is the whole flow.
How much can I submit?¶
Every approved account starts on the academic tier: 5 jobs a day, 30 a month, and 50,000 tiles a day.
Limits states every counter, cap and rate.
Can I analyse a cohort?¶
Yes, one job per slide. Limits says how to ask for more than your counters allow.
Your results¶
What is in a result file?¶
One .h5ad per job: one row per tile, one column per gene, your coordinates
beside them. What you get back says what you can explore in it.
Do I need a bulk RNA profile?¶
No.
Add the sample's bulk RNA profile to the prediction for its slide, and keep the same genes and file format. Without a bulk profile, Aurora predicts from the slide alone.
Bulk RNA shows the two calls.
How do I cite Aurora?¶
Cite the DeepSpot-M preprint in any work that uses a result.
Can I use a result to decide patient care?¶
No.
Responsible use states what a result may be used for.
Technical help¶
How do I find out what went wrong?¶
Every failure carries a stable code and a request identifier.
Errors says what each code means and what to change.