Downloaing SRA using Aspera

Question

[always_learning]

Dear Bioinformaticians,

I was trying to download some data from SRS using Aspera.

I was able to download this example SRR304976 mentioned at http://www.ncbi.nlm.nih.gov/books/NBK158899/ sample

ascp -T -l640M -i ~/.aspera/connect/etc/asperaweb_id_dsa.openssh  [e-mail]:/sra/sra-instant/reads/ByRun/sra/SRR/SRR304/SRR304976/SRR304976.sra .

But when I am trying to download sample SRR2125414 http://www.ncbi.nlm.nih.gov/Traces/study/?acc=SRP060765%2CSRP061943%2CSRP061463&go=go using below command

ascp -T -l640M -i ~/.aspera/connect/etc/asperaweb_id_dsa.openssh  [e-mail]:/sra/sra-instant/reads/ByRun/sra/SRR/SRR212/SRR2125414/SRR2125414.sra .

Then I am getting below error.

Session Stop  (Error: Server aborted session: No such file or directory)

If someone else has seen same problem and than what was workaround? Please guide me on this problems.

Thanks

[GenoMax]

If you look here, there is no .sra file in that directory. In fact there are no files in that directory at all.

[always_learning]

So what will be folder structure than ? as per example given at that page, I have given correct structure.

Thanks

[GenoMax]

Is this a controlled access sample? Corresponding page at ENA also has no sequence files either.

[always_learning]

I can able to download same data using fastq-dump SRR2125414 and can access using fastq-sam SRR2125414 as well.

Answer 1

[always_learning]

I wrote to NCBI and they replied on this issue. I am posting here for others now.

Hi,

Thanks for writing to us.

To access files through Aspera or ASCP, the request goes through a separate route. For this specific subset, the web-equivalent venue is this: http://www.ncbi.nlm.nih.gov/public/?/sra/sra-instant/reads/ByRun/sra/SRR/SRR212/

If you examine the directories available, you can see that the ones for this set of files is not there. The regular ftp does have this listed: ftp://ftp.ncbi.nlm.nih.gov/sra/sra-instant/reads/ByRun/sra/SRR/SRR212/

The issue is that ASCP access requires special indexing of the available content and from what I can see that indexing process is lagging behind. I have contacted our developers over this. They will update this as soon as they can. Your patience will be greatly appreciated.

You will need to use regular ftp, wget, or prefetch (from ncbi sratoolkit and is preferred by sra group) to get the data set. The latter is much easier: prefetch SRR2125414 SRR2125415 SRR2125413 SRR2125412

This also revealed a feature unknown to me: prefetch is smarter or have better access to internal ncbi file structure to locate the files AND to use the Aspera/ascp set up I have. This is an added reason for you to switch to use prefetch. You will need to tools to extract the data after all.

Regards,