GATK 4.4.0 Conda Environment Installation Error: ResolvePackageNotFound: - defaults::tensorflow=1.15.0
Hello,
I am attempting to run the CNNScoreVariants command on an HPC system. I followed the instructions to create a conda environment posted here: https://gatk.broadinstitute.org/hc/en-us/articles/360035889851--How-to-Install-and-use-Conda-for-GATK4.
I am using GATK version 4.4.0. Below is the exact command I used:
conda env create --prefix PATH/TO/conda_envs/GATK -f gatkcondaenv.yml
c) Entire program log:
Collecting package metadata (repodata.json): done
Solving environment: failed
ResolvePackageNotFound:
- defaults::intel-openmp=2019.4
- defaults::tensorflow=1.15.0
ATTEMPT TO FIX:
I was able to overcome the intel-openmp failure by adding anaconda to the channels list in the yml file and then changing -defaults to -anaconda::intel-openmp=2019.4.
I am unable to find an alternate channel that has tensorflow 1.15.0. Owing to the python version specified in the environment, I am unable to upgrade tensorflow. I have been able to install 1.10, but the analysis fails with the following:
[0m15:41:10.379 INFO NativeLibraryLoader - Loading libgkl_compression.so from jar:file:/project/software/el9/apps/gatk/4.4.0.0/gatk-package-4.4.0.0-local.jar!/com/intel/gkl/native/libgkl_compression.so
15:41:10.493 INFO CNNScoreVariants - ------------------------------------------------------------
15:41:10.498 INFO CNNScoreVariants - The Genome Analysis Toolkit (GATK) v4.4.0.0
15:41:10.500 INFO CNNScoreVariants - For support and documentation go to https://software.broadinstitute.org/gatk/
15:41:10.502 INFO CNNScoreVariants - Executing as stephen.harding@ceres20-compute-6.scinet.local on Linux v5.14.0-284.30.1.el9_2.x86_64 amd64
15:41:10.505 INFO CNNScoreVariants - Java runtime: Java HotSpot(TM) 64-Bit Server VM v17.0.6+9-LTS-190
15:41:10.507 INFO CNNScoreVariants - Start Date/Time: October 26, 2023 at 3:41:10 PM CDT
15:41:10.509 INFO CNNScoreVariants - ------------------------------------------------------------
15:41:10.511 INFO CNNScoreVariants - ------------------------------------------------------------
15:41:10.516 INFO CNNScoreVariants - HTSJDK Version: 3.0.5
15:41:10.518 INFO CNNScoreVariants - Picard Version: 3.0.0
15:41:10.522 INFO CNNScoreVariants - Built for Spark Version: 3.3.1
15:41:10.524 INFO CNNScoreVariants - HTSJDK Defaults.COMPRESSION_LEVEL : 2
15:41:10.526 INFO CNNScoreVariants - HTSJDK Defaults.USE_ASYNC_IO_READ_FOR_SAMTOOLS : false
15:41:10.530 INFO CNNScoreVariants - HTSJDK Defaults.USE_ASYNC_IO_WRITE_FOR_SAMTOOLS : true
15:41:10.532 INFO CNNScoreVariants - HTSJDK Defaults.USE_ASYNC_IO_WRITE_FOR_TRIBBLE : false
15:41:10.534 INFO CNNScoreVariants - Deflater: IntelDeflater
15:41:10.537 INFO CNNScoreVariants - Inflater: IntelInflater
15:41:10.539 INFO CNNScoreVariants - GCS max retries/reopens: 20
15:41:10.541 INFO CNNScoreVariants - Requester pays: disabled
15:41:10.544 INFO CNNScoreVariants - Initializing engine
15:41:10.858 INFO FeatureManager - Using codec VCFCodec to read file file:///project/mpm_fusarium_genomics/stephen/GATK_Test/Fpro10717_TrimmedPair_OutFiles/Fpro10717.g.vcf.gz
15:41:10.964 INFO CNNScoreVariants - Done initializing engine
15:41:10.968 INFO NativeLibraryLoader - Loading libgkl_utils.so from jar:file:/project/software/el9/apps/gatk/4.4.0.0/gatk-package-4.4.0.0-local.jar!/com/intel/gkl/native/libgkl_utils.so
15:41:32.284 INFO CNNScoreVariants - Using key:CNN_1D for CNN architecture:/tmp/1d_cnn_mix_train_full_bn.693313877541357885.json and weights:/tmp/1d_cnn_mix_train_full_bn.8688214035610729398.hd5
15:41:32.297 INFO CNNScoreVariants - Done scoring variants with CNN.
15:41:32.300 INFO CNNScoreVariants - Shutting down engine
[October 26, 2023 at 3:41:32 PM CDT] org.broadinstitute.hellbender.tools.walkers.vqsr.CNNScoreVariants done. Elapsed time: 0.37 minutes.
Runtime.totalMemory()=285212672
org.broadinstitute.hellbender.utils.python.PythonScriptExecutorException: A nack was received from the Python process (most likely caused by a raised exception caused by): nkm received
: Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/project/mpm_fusarium_genomics/stephen/conda_envs/GATK/lib/python3.6/site-packages/vqsr_cnn/vqsr_cnn/models.py", line 21, in start_session_get_args_and_model
K.clear_session()
AttributeError: module 'keras.backend' has no attribute 'clear_session'
at org.broadinstitute.hellbender.utils.python.StreamingPythonScriptExecutor.waitForAck(StreamingPythonScriptExecutor.java:222)
at org.broadinstitute.hellbender.utils.python.StreamingPythonScriptExecutor.sendSynchronousCommand(StreamingPythonScriptExecutor.java:183)
at org.broadinstitute.hellbender.tools.walkers.vqsr.CNNScoreVariants.initializePythonArgsAndModel(CNNScoreVariants.java:559)
at org.broadinstitute.hellbender.tools.walkers.vqsr.CNNScoreVariants.onTraversalStart(CNNScoreVariants.java:319)
at org.broadinstitute.hellbender.engine.GATKTool.doWork(GATKTool.java:1096)
at org.broadinstitute.hellbender.cmdline.CommandLineProgram.runTool(CommandLineProgram.java:149)
at org.broadinstitute.hellbender.cmdline.CommandLineProgram.instanceMainPostParseArgs(CommandLineProgram.java:198)
at org.broadinstitute.hellbender.cmdline.CommandLineProgram.instanceMain(CommandLineProgram.java:217)
at org.broadinstitute.hellbender.Main.runCommandLineProgram(Main.java:160)
at org.broadinstitute.hellbender.Main.mainEntry(Main.java:203)
at org.broadinstitute.hellbender.Main.main(Main.java:289)
Using GATK jar /project/software/el9/apps/gatk/4.4.0.0/gatk-package-4.4.0.0-local.jar
Running:
java -Dsamjdk.use_async_io_read_samtools=false -Dsamjdk.use_async_io_write_samtools=true -Dsamjdk.use_async_io_write_tribble=false -Dsamjdk.compression_level=2 -jar /project/software/el9/apps/gatk/4.4.0.0/gatk-package-4.4.0.0-local.jar CNNScoreVariants -V Fpro10717_TrimmedPair_OutFiles/Fpro10717.g.vcf.gz -R F_proliferatum_62905_clean_sorted_AddChrInfo.fa -O Fpro10717_TrimmedPair_OutFiles/Fpro10717.annotated.vcf
-
Can you try updating your conda installation via conda update conda in the base environment? It may be possible that an older conda installation might not communicate properly with the repository. If this does not solve your issue we also recommend using the docker image which already has all the dependencies built-in.
I hope this helps.
Regards.
Please sign in to leave a comment.
1 comment