Creating a Simple Example Database

Creating a simple database with simulated data provides a convenient way to test whether you are able to run the PHG software successfully. A simple reference genome along with fastq files of a few taxa can be created using the CreateSmallGenomesPlugin. The database can be created either using the PHG Docker or using the standalone version of TASSEL. The following TASSEL command can be used to create the data after replacing "/path/to/basedir/" with a directory name.

./run_pipeline.pl -debug -Xmx1G -CreateSmallGenomesPlugin \
   -baseDir /path/to/basedir -geneLength 3000 -interGeneLength 3500 \
   -numberOfGenes 10 -endPlugin

The command can be run without the -baseDir parameter, in which case the baseDir default will be userDir/temp/phgSmallSeq. If you use docker to run the command (below), userDir = (root). In that case using the default value, the data will be written to root not to /phg/ and the data will disappear when the container is removed. In short, when running CreateSmallGenomesPlugin using the docker always use the -baseDir parameter and start it with /phg/.

When using a docker to run the command, a directory called "basedir" will be created in the directory specified by "/Path/to" if it does not already exist. To run the command using the docker

docker run --name smallGenomes_container --rm -v /Path/to/:/phg/ -t maizegenetics/phg \
   /tassel-5-standalone/run_pipeline.pl -debug -Xmx1G \
   -CreateSmallGenomesPlugin -baseDir /phg/basedir \
   -geneLength 3000 -interGeneLength 3500 \
   -numberOfGenes 10 -endPlugin > /path/to/smallGenomes_log.txt

The plugin runs MakeDefaultDirectoryPlugin and populates some of the resulting directories with simulated data representing a single chromosome with 10 (-numberOfGenes) genes. Specifically, it creates two sub-directories, "answer" and "dockerBaseDir" in the folder specified by the -baseDir parameter. The "answer" directory holds data that can be used to validate imputation results from running the PHG. The "dockerBaseDir" directory holds data that can be used to populate the PHG, map reads, and impute data. It will also hold files produced by running PHG. For the docker commands that follow, "dockerBaseDir" is the directory that should be mounted as "/phg/".

Using the Example Database

The four pipeline plugins used for the PHG are

If you ran CreateSmallGenomesPlugin one of the things it did was to run MakeDefaultDirectoryPlugin so that is already done. Following are the docker commands and links to config files needed to run the other plugins. The first step of each pipeline is to make a call to liquibase to make sure that the version of the PHG software and the PHG database are the same. As a result you need to have pulled the same versions of both maizegenetics/phg and maizegenetics/phg_liquibase.

MakeInitialPHGDBPipelinePlugin

The config file needed here was written to dockerBaseDir when CreateSmallGenomesPlugin was run. The docker command to run this pipeline is

docker run --name makeDb_container --rm -v /path1/to/dockerBaseDir:/phg/ -t maizegenetics/phg \
   /tassel-5-standalone/run_pipeline.pl -debug -Xmx1G -configParameters /phg/configSQLiteDocker.txt \
   -MakeInitialPHGDBPipelinePlugin -endPlugin > /path2/to/makedb_log.txt

Replace "/path/to" with the path that is correct for your computer. The path to the log file is on your computer and outside of docker. "maizegenetics/phg" is the name of the docker image. If the plugin ran correctly, "phgSmallSeq.db" will have been created in dockerBaseDir. Also, you should look at the log file to make sure there are no error messages.

PopulatePHGDBPipelinePlugin

The docker command to run this pipeline, after replacing both instances of "/path/to", is

docker run --name populateDb_container --rm -v /path/to/dockerBaseDir:/phg/ \
   -t maizegenetics/phg /tassel-5-standalone/run_pipeline.pl -debug -Xmx1G \
   -configParameters /phg/configSQLiteDocker.txt -PopulatePHGDBPipelinePlugin -endPlugin \
   > /path/to/populatedb_log.txt

This step may take several minutes. It is aligning fastq files to reference and extracting WGS based haplotypes using the script CreateHaplotypesFromFastq.groovy. It will populate the database with haplotypes and write some files to folders in dockerBaseDir in the process. As usual, check the log file to make sure there were no errors and that the plugin ran as expected.

ImputePipelinePlugin

The config file needed for this step is imputeConfig.txt. You will need to put a copy of that in dockerBaseDir. The docker command to run this pipeline, after replacing both instances of "/path/to" and copying the config file to baseDockerDir, is

docker run --name impute_container --rm -v /path/to/dockerBaseDir:/phg/ -t maizegenetics/phg \
   /tassel-5-standalone/run_pipeline.pl -debug -Xmx1G -configParameters /phg/imputeConfig.txt \
   -ImputePipelinePlugin -imputeTarget pathToVCF -endPlugin > /path/to/impute_log.txt

The final result is a vcf file, dockerBaseDir/outputDir/output.vcf.

Examine the data

An excellent way to look at the database contents is to use rPHG.

Wiki

PracticalHaplotypeGraph / UserInstructions / ExampleDatabase