Updated 2020-04-03 by @samstudio8

So you think you are ready to provide metadata

You have read the docs?

You know what a biosample, library and sequencing run is and what identifier they should have (ie. you read Terminology and minimal metadata)
If you’re submitting via the web tool, you have read those docs?

You have provided at least the minimum metadata?
Your samples have one and only one, single, unique, shareable identifier?
You know that the “COGUK ID” does not necessarily have to be a WSI “Heron” barcode, but that it should be consistent and is usually prefixed by some short code for your site?
Your pooled sample libraries have a name that is reasonably unique? (BIRM-LIBRARY-001 good, 1 bad).
Your sequencing runs have a name that is reasonably unique? (<date>_<machine_id>_<run_no>_<some_zeros>_<flowcell> good, my-run bad)
You are sure that you have permission to provide us this metadata?
You are sure that any identifiers are not personally identifiable?
If multiple samples have come from the same patient, you have set the biosample_source_id to the first COGUK sample identifier for all of the related samples?
If you are re-sequencing a sample from another site, you have not changed the central_sample_id?
You are aware that once a record has been submitted, its central_sample_id (COGUK ID), library_name and run_name cannot be changed. If you were thinking about changing the COGUK ID, it is now or never.
For spreadsheet-like uploads, your data is in a CSV or XLSX with the right template for the webtool?
It seemed to work when youn ran it on the test version of Majora? (See below)

You have registered for a test account?
You know that this test account is different from your real CLIMB account and begins with test-?
You know that the testing version of Majora has a magenta coloured banner at the top to remind you that it’s the test version?
You have your API key token from your test profile?
You have tried one of the two ways to upload metadata and checked it looked right on the website?

You have downloaded ocarina
You know that the configration is saved to ~/.ocarina by default and instructions for doing this appear the first time you run the tool?
You have set the MAJORA_DOMAIN to https://majora-test.covid19.climb.ac.uk/ for testing?
You set the MAJORA_USER and MAJORA_TOKEN to your testing username and token?
You have tried uploading data with:
- ocarina put biosample?
- ocarina put library?
- ocarina put sequencing?
So you know that for ocarina uploads, you have to provide a biosample*, then a library, then the sequencing run, in that order? (Ask @samstudio8 about generating dummy biosamples if you want to skip straight to libraries and sequencing…).

You registered for a real unified CLIMB account and you’ve had an e-mail to tell you it was approved?
You still know your password? (Reset)
You have an API key token on your CLIMB profile?
You know that this account is your real account because the username does not start with test-?
You know that the API key for this account is different from the one for your test account?
You know that this username and API key will only work on the real Majora website?
For webtool uploads: provide your real username and token and upload as normal
For ocarina uploads: set MAJORA_DOMAIN to https://majora.covid19.climb.ac.uk/ and update the MAJORA_USER and MAJORA_TOKEN to your real credentials?
You promise to be careful to not submit test data to the real database because it’s really hard to get rid of data once it is there?
You know that we know who has uploaded bad metadata?
You know that without sufficient metadata, CLIMB will not publish your sample or consensus?

Published 2020-04-03. Updated 2020-04-03. Page maintainer @samstudio8.