all the deposited samples from "Seafloor Basalts" paper seem to be the same (!) about astrobiomike.github.io

[AstrobioMike]

Hi there!

Sorry you've been messing with this trying to figure it out! It is messed up. Each one of those holds ALL of the samples mixed together. So you can just pull any one of them, and then demultiplex that one into all the samples. The biosample data holds the barcodes for each individual sample, e.g. blank-4 is GAGTTGAG. I screwed up uploading this, was my first time :) I tried to sort it out a while back and gave up. You can get them all in a file from that page like so:

Then some bash magic will turn that into a cleaner mapping file, here's one kinda ugly example if you're newer to that side of things :)

paste <(grep "Identifiers" biosample_result.txt | cut -f 4 -d ":" | cut -f 1 -d ";" | cut -c 2-) <(grep "barcode" biosample_result.txt | sed 's/^.*=//' | tr -d '"') | tr -d '*' > map.tsv

[AstrobioMike]

Actually you look pretty comfortable with bash already, so forget all that and just take the apology for the annoying archived data :/

[matrs]

I see, I was thinking about the size of each archive, I don't have much experience with amplicon data obviously but I had seen way smaller archives before. I downloaded the files using cli clients and as soon as i got them I run cutadapt and other few steps so I didn't catch this "detail". It seems that with databases and everything on the internet is the same, once you uploaded something, you cannot do a single thing to erase it ;].
Thanks for your very quick answer and for the free quality content you put on your website.