Open mcourtot opened 4 years ago
@mcourtot
I have understood all the other tickets except this. I will quickly chat with you on the standup tomorrow on this ticket.
the documentation for a run says "Run This section enables you to describe the experimental metadata surrounding the runs. All mandatory fields should be completed, including a descriptive experiment name, sequencing platform, sequencing instrument, library source, library selection and library strategy. This also includes the name of the file (to be uploaded) and also its md5 checksum value. For guidance in this, see documentation here. An optional description of the library and its name can be provided, e.g. a link to a specific protocol that was used to obtain the data."
This seems to say that the md5 checksum is a mandatory field for metadata submission, while we discussed not making this compulsory, which is why we are computing it on the fly upon submission.
This may be a question for Nadim and Jeena - should the spreadsheet enforce submission of the md5 or not - and then update the documentation accordingly.
md5 shouldn't be included. Sorry, this wasn't edited after yesterday's call. Confirming the md5 should not be included in the metadata file as agreed in yesterday's call. The doc says not to edit but the change should be: "This also includes the name of the file (to be uploaded) and also its md5 checksum value. For guidance in this, see documentation here." shortened to "This also includes the name of the file to be uploaded"
Run documentation states that md5 should be provided in the file.
If we make it mandatory in the metadata spreadsheet why do we need to regenerate it ourselves?