STAR not working with index in an S3 bucket

Question

[jeremy]

Hello, I was wondering if anybody had experience using STAR with AWS. I have no problems running it locally, but am now transitioning it for use with snakemake/tibanna framework. With snakemake/tibanna, all the files you work with must be in an S3 bucket on AWS. So I put my genome index (that works locally for alignment) in an S3 bucket, referenced as such

S3.remote("rnaseq.raw/indexes/STAR_index"),

However, when I try this, I keep encountering the error

EXITING because of FATAL ERROR: could not open genome file rnaseq.raw/indexes/STAR_index/genomeParameters.txt

However, this file is definitely there in that folder. I can't find anything in the STAR docs about working with remote files. Does anybody have any experience with this?

[GenoMax]

Can you clarify if this question is about using STAR on AWS or about using STAR locally but with index in an S3 bucket? This may be a case of authentication error since the program is not able to open the file.