Cannot download BAMs from SRA AWS

Question

[devenvyas]

I am trying to get the BAM files from these SRA accessions

https://trace.ncbi.nlm.nih.gov/Traces/sra/?run=SRR1187911 https://trace.ncbi.nlm.nih.gov/Traces/sra/?run=SRR1045127

I have tried to download these both to our lab server and Stony Brook's cluster, and I get these same errors

[Vyas@sahara test]$ aws s3 ls --no-sign-request s3://sra-pub-src-5/SRR1187911

An error occurred (AccessDenied) when calling the ListObjectsV2 operation: Access Denied
[Vyas@sahara test]$ aws s3 ls --no-sign-request s3://sra-pub-src-10/SRR1045127

An error occurred (AccessDenied) when calling the ListObjectsV2 operation: Access Denied

Not sure what is going on... Anyone have an idea?

Answer 1

[GenoMax]

Looks like there is no free egress for these BAM files. You will need to have a aws/google account that you can pay with to download.

400 Bucket is a requester pays bucket but no user project provided.

[keller.mckowen]

I can't access them on the AWS CLI even with credentials. I don't think this is a public bucket.

However, I can download via http. (Why is this open but AWS is locked?) So I could do it that way and upload to my S3, but this is super inefficient.

The other option is the SRA Data Delivery service. Edit: This service is actually pretty easy to use and fast. You do need an NCBI account, and have to pay data transfer fees.

Are there some special permissions I need to access these sra-pub-scr-(>2) buckets?

Answer 2

[Raman2]

Using fasq-dump or fasterq-dump (https://github.com/ncbi/sra-tools/tree/master/tools/fasterq-dump) from sra-tools would be the easiest way to download files from SRA

[GenoMax]

It is not a question of the tools. I don't think it is possible to download "original format" data (in this case BAM files that are available under the "Data Access" tab) using the tools you mention.