Closed lingnanyuan closed 1 year ago
Hi,
Seems like you are running as sudo? what is the reason for that? I could imagine that this can lead to some libPath issues. Furthermore, I can see it seems like you are running this on a laptop? The RAM limit you set will not be followed by zUMIs - especially since loading the human genome for mapping using STAR needs at least approx. 25 GB RAM
If I don't run sudo, it will say
YAML file has an error. Look at the zUMIs_YAMLerror.log or contact developers.
The log attached: Spike.zUMIs_YAMLerror.log Basically cannot detect the fastq files.
And the yaml file: Spike.txt
I'm testing on one chromosome so I think it would not take that many RAM.
Hi Lingnan,
The error log says the path to the barcode file is not found. Please double check that. Depending on the number of reads, mapping against only one chromosome can be quite slow because STAR will have trouble finding alignments.
Best, Christoph
Hi Christoph,
Thanks again for your patience. There might be something wrong with my environment, if I run with sudo, it cannot find the r packages, if I run without sudo, it cannot detect the fastq files. Maybe I'll try with other device latter.
I dont recommend usage of sudo, and if you are in doubt with the installation of dependencies, please use the built-in conda environment (zUMIs.sh -c -y config.yaml)
bash ~/Desktop/zUMIs/zUMIs/zUMIs.sh -c -y Spike.yaml
Using miniconda environment for zUMIs!
note: internal executables will be used instead of those specified in the YAML file!
/home/lynn/Desktop/zUMIs/zUMIs/zUMIs.sh: line 161: /home/lynn/Desktop/zUMIs/zUMIs/zUMIs-env/bin/activate: No such file or directory
/home/lynn/Desktop/zUMIs/zUMIs/zUMIs.sh: line 162: conda-unpack: command not found
YAML file has an error. Look at the zUMIs_YAMLerror.log or contact developers.
The error log is the same as no fastq file detected.
Hi Lingnan,
The first time you run zUMIs after cloning from GitHub, it is critical not to abort the initial unpacking process that will run automatically. This is the reason for your errors eg. "conda-unpack: command not found".
You should simply delete your zUMIs installation and clone it new from GitHub.
Best, Christoph
I'm trying to work with zUMIs with some yaml file. I have the R 4.1.1 installed and all the packages installed as well. But when I try to run zUMIs.sh, it reports:
To Reproduce yaml file attached: Spike.txt
Desktop (please complete the following information): Ubuntu 20.04.4 LTS
Thank you in advance! Lingnan