Getting biological replicates from SRA run NCBI

Question

[Andres]

I want to download the paired RNA-seq libraries from this experiment SRX2613389. There are 8 SRA runs that from this article correspond to 8 tissues, each tissue has 3 biological replicates(they took samples from 3 different plants per tissue) which would result in 24 libraries in total. I know how to get the .sra files and then the Forward and reverse reads. But i don't know how to get the biological replicates that seem to be concatenated on each of the .sra files. If someone knows how to split that i would love to know.

Thanks, have a nice day.

PD: In the SRA run browser of each accession in "data access" i can see the individual files for each replicate under the "originally submitted data". I followed tutorials on how to access that data with an Amazon web services(AWS) account but i haven't been able to access the data.

PD2: In one of the additional files of the article exists a table with the read count of each individual library, ¿would it be reasonable to simply split each sra file by the number of reads each replicate has?, assuming they concatenated the files with a simple cat command.

[GenoMax]

Download the fastq data from EBI-ENA for this experiment.

[Andres]

I only see there the fastq files that i already have, i want to know if theres a way to split those into the biological replicates each file has.

[GenoMax]

would it be reasonable to simply split each sra file by the number of reads each replicate has?

You could write to the submitter's and ask if that is what they did but then you don't know if they used a certain order to cat the data so too many unknowns. Data does not appear to dump out with original illumina headers either.

but i haven't been able to access the data.

That is because cloud bucket's are set to requester pays that means you will need to pay for the download. If you are able/willing then this is an option.

You may want to consider writing to SRA support desk and ask them about this case where replicates are available via cloud payments but not via free downloads. It would be interesting see what they say. Moving the data to cloud always had the potential of locking some users out (who can't pay) and this appears to be one example of that.

[Andres]

Sorry for not answering before, i forgot. But just if anyone is interested, I ended up paying for an amazon bucket, the cost was like 10 bucks, for like 20Gb of data movement, was a bit annoying but i was able to get the metadata.

Thank you for your insight.