klymenko
commented
10 months ago Most likely you do not have access to NCBI servers.
Closed m-neuhoff closed 10 months ago
klymenko
commented
10 months ago Most likely you do not have access to NCBI servers.
Hi, I have installed the sratoolkit sticking to this instructions https://github.com/ncbi/sra-tools/wiki but it is
1. not downloading any files
2. nor creating any directories
3. and unfortunately also not giving me error messages to work on.
I am working on a cluster with slurm as workload manager. If anyone could point me into a direction, whether it is the internet connection or something else, share a helpful link or a tip what I forgot to check or do, everything is welcome. =)
I used this commands: 1) /path/to/prefetch --max-size 10g --option-file /path/to/SRR_test_list.txt 2) /path/to/prefetch --max-size 10g SRR11963181
I got this in my errorfile for 1) (list contains two SRR_IDs): 2023-11-23T05:40:30 prefetch.3.0.7: Current preference is set to retrieve SRA Lite files with simplified base quality scores. 2023-11-23T05:41:30 prefetch.3.0.7: Current preference is set to retrieve SRA Lite files with simplified base quality scores. this for 2): 2023-11-23T10:11:57 prefetch.3.0.7: Current preference is set to retrieve SRA Lite files with simplified base quality scores. (accordingly for normal quality)
I tried using srapath in the script to check wether it can find a path too, but it does not produce any output nor error message. If I use the same command directly in the terminal, it does print the correct link, though.
What I did or checked to get prefetch working: In the sratoolkit configuration remote access and local file caching is enabled. I set the user repository and ticked the box that it saves downloaded files in the user repository. I defined the error and output files to be created in a different directory than the one I defined as the user repository. I checked the available space. I set the --max-size so that it is enough for my test run (with only one to two files to download). I tried with lite and with normal quality. I tried different SRR-IDs. I scaled up the memory request for the slurm job.