_installGATK issues

[ssmith151]

Error:

/bin/sh: gatk-register: command not found
CalledProcessError in line 26 of /home/ss15w/Software/MAMBA/MAMBA.snakefile:
Command 'gatk-register /home/ss15w/Software/GenomeAnalysisTK-.tar' returned non-zero exit status 127
File "/home/ss15w/Software/MAMBA/MAMBA.snakefile", line 26, in
File "/home/ss15w/Software/MAMBA/scripts/utils/config_setup.py", line 21, in updateConfig
File "/home/ss15w/Software/MAMBA/scripts/utils/config_setup.py", line 34, in _checkForGATK
File "/home/ss15w/Software/MAMBA/scripts/utils/config_setup.py", line 47, in _installGATK
File "/home/ss15w/anaconda3/envs/MAMBA/lib/python3.5/subprocess.py", line 316, in check_output
File "/home/ss15w/anaconda3/envs/MAMBA/lib/python3.5/subprocess.py", line 398, in run
Registering GATK:

Received Error with all versions of Gatk: GenomeAnalysisTK-.tar, GenomeAnalysisTK.jar, and GenomeAnalysisTK-.tar.bz2 as options in config.yaml

Please provide an example of getting config.yaml to recognize gatk.

[vangalamaheshh]

@ssmith151 , you just have to mention the path of GATK in config.yaml. Check "gatk_exec" option in config.yaml.

• Mahesh

[ssmith151]

The path to GATK has been used in config.yaml by the gatk_exec option in the three different formats with this same error.

Do you have an example of the config.yaml with an example of the path to gatk for reference?

[vangalamaheshh]

@ssmith151 , /home/ss15w/Software/GenomeAnalysisTK-.tar has to be a bzipped tar. I notice in README, I mentioned "bunzipped tar". That's a typo. You don't have to bunzip that tar you downloaded from GATK. I will go ahead and edit README to reflect this change.

Let me know if you still have issues.

Thanks,
Mahesh

[vangalamaheshh]

@ssmith151 , as far as this error goes,

/bin/sh: gatk-register: command not found

I am not sure why you are seeing this error. Once you install MAMBA environments, you should have this dependency installed in your /path/to/miniconda3/envs/MAMBA/bin/gatk-register.

Let me know.

• Mahesh

[ssmith151]

removed the environment and re-added it.

Verified that gatk-register was in the bin of the MAMBA environment.

New error within the same command:

First:

Registering GATK:
Error: Could not create the Java Virtual Machine.
Error: A fatal exception has occurred. Program will exit.
CalledProcessError in line 26 of /home/ss15w/Software/MAMBA/MAMBA.snakefile:
Command 'gatk-register /home/ss15w/Software/GenomeAnalysisTK-.tar.bz2' returned non-zero exit status 1
File "/home/ss15w/Software/MAMBA/MAMBA.snakefile", line 26, in
File "/home/ss15w/Software/MAMBA/scripts/utils/config_setup.py", line 21, in updateConfig
File "/home/ss15w/Software/MAMBA/scripts/utils/config_setup.py", line 34, in _checkForGATK
File "/home/ss15w/Software/MAMBA/scripts/utils/config_setup.py", line 47, in _installGATK
File "/home/ss15w/anaconda3/envs/MAMBA/lib/python3.5/subprocess.py", line 316, in check_output
File "/home/ss15w/anaconda3/envs/MAMBA/lib/python3.5/subprocess.py", line 398, in run

After this:

mkdir: cannot create directory `/tmp/gatk': File exists
Error: Could not create the Java Virtual Machine.
Error: A fatal exception has occurred. Program will exit.
CalledProcessError in line 26 of /home/ss15w/Software/MAMBA/MAMBA.snakefile:
Command 'gatk-register /home/ss15w/Software/GenomeAnalysisTK-.tar.bz2' returned non-zero exit status 1
File "/home/ss15w/Software/MAMBA/MAMBA.snakefile", line 26, in
File "/home/ss15w/Software/MAMBA/scripts/utils/config_setup.py", line 21, in updateConfig
File "/home/ss15w/Software/MAMBA/scripts/utils/config_setup.py", line 34, in _checkForGATK
File "/home/ss15w/Software/MAMBA/scripts/utils/config_setup.py", line 47, in _installGATK
File "/home/ss15w/anaconda3/envs/MAMBA/lib/python3.5/subprocess.py", line 316, in check_output
File "/home/ss15w/anaconda3/envs/MAMBA/lib/python3.5/subprocess.py", line 398, in run
Registering GATK:

[vangalamaheshh]

It seems that's your compute specific issue.
I have seen this error on a cluster due to admin set up of how Java is set up.
If you are on a cluster, try to launch a cluster job within interactive queue and execute this command.

[ssmith151]

Possibly. This command is being run on the cluster.

Using -q interactive and removing the gatk from /tmp/ :

mkdir: cannot create directory `/tmp/gatk': File exists
Error: Could not create the Java Virtual Machine.
Error: A fatal exception has occurred. Program will exit.
CalledProcessError in line 26 of /home/ss15w/Software/MAMBA/MAMBA.snakefile:
Command 'gatk-register /home/ss15w/Software/GenomeAnalysisTK-.tar.bz2' returned non-zero exit status 1
File "/home/ss15w/Software/MAMBA/MAMBA.snakefile", line 26, in
File "/home/ss15w/Software/MAMBA/scripts/utils/config_setup.py", line 21, in updateConfig
File "/home/ss15w/Software/MAMBA/scripts/utils/config_setup.py", line 34, in _checkForGATK
File "/home/ss15w/Software/MAMBA/scripts/utils/config_setup.py", line 47, in _installGATK
File "/home/ss15w/anaconda3/envs/MAMBA/lib/python3.5/subprocess.py", line 316, in check_output
File "/home/ss15w/anaconda3/envs/MAMBA/lib/python3.5/subprocess.py", line 398, in run
Registering GATK:

[vangalamaheshh]

I suggest you install MAMBA and run it with test data following README using a linux/unix based desktop first. That will help you get started with out getting bogged down with cluster set up and what not.

[vangalamaheshh]

Also, the error message mkdir: cannot create directory `/tmp/gatk': File exists is strange. May I know what is the command that you are trying to run. Paste your config.yaml and let me know the commands you are running to see the error message that you are seeing. That will be of help to understand what's going on.

[ssmith151]

Config.yaml

metasheet: meta.csv
min_mem: 5000
med_mem: 15000
max_mem: 30000
min_cores: 1
med_cores: 2
max_cores: 12
read_depth: null
map_quality: null
kingdom: null
genus: Staphylococcus
species: aureus
strain: null
gram: null
gatk_exec: /home/ss15w/Software/GenomeAnalysisTK-.tar.bz2
reference: null
isolates:
159_BE_8:

• MAMBA_TEST_DATA/159_BE_8_R1.fastq.gz
• MAMBA_TEST_DATA/159_BE_8_R2.fastq.gz
  34_SE_45:
• MAMBA_TEST_DATA/34_SE_45_R1.fastq.gz
• MAMBA_TEST_DATA/34_SE_45_R2.fastq.gz
  47_BE_45:
• MAMBA_TEST_DATA/47_BE_45_R1.fastq.gz
• MAMBA_TEST_DATA/47_BE_45_R2.fastq.gz
  246_DE_225:
• MAMBA_TEST_DATA/246_DE_225_R1.fastq.gz
• MAMBA_TEST_DATA/246_DE_225_R2.fastq.gz
  263_DE_9:
• MAMBA_TEST_DATA/263_DE_9_R1.fastq.gz
• MAMBA_TEST_DATA/263_DE_9_R2.fastq.gz

Command:

snakemake -s ./MAMBA.snakefile --cluster 'bsub -n 2 -J MAMBA -W 360 -q interactive -R rusage[mem=4000]' --jobs 2 --latency-wait 60 >&MAMBA.log

Using PuTTY interface, for job creation so the double quotes are single quotes to get the bsub to work. Please let me know if the command is being input incorrectly. It may be helpful to have an example command for LSF interfaces withing the demo set.

[vangalamaheshh]

@ssmith151 , OK, on LSF cluster, do this,

bsub -I -q interactive bash
snakemake -s MAMBA/MAMBA.snakefile -n
exit

-n flag makes sure the pipeline is not launched. But will list the rules that will be executed along with installing GATK.

Now, since, you can't run java on master node, edit config.yaml and add a param,

skip_gatk_check: true

Now you can launch pipeline listed on README LSF cluster section. Copy the exact command (with out any modifications) should run the pipeline.

Let me know.

• Mahesh

[vangalamaheshh]

Sorry, the only things you need to change are,

-q <queue_name>
--jobs

• Mahesh

[vangalamaheshh]

@ssmith151 , I have added video link on how to install MAMBA on project home page: https://dst-umms.github.io/MAMBA/ . Let me know if you still have issues.

• Mahesh