Omics Pipe Tutorial – Configuring the Parameter File¶
Before running Omics Pipe, you must configure the parameters file, which is a YAML document. Example parameters files are located within the omics_pipe/test folder for each pipeline. Copy one of these parameters files into your working directory, and edit the parameters to work with your sample names, directory structure, software options and software versions. Make sure to keep the formatting and parameter names exactly the same as in the example parameters files.
Note
Make sure to follow the YAML format exactly. Ensure that there is only one space after each colon.
Note
For parameters in quotes in the test parameters file, please make sure to keep them in quotes in your custom parameter file.
The STEP parameter should be the function name of the last step in the pipeline that you want to run (e.g. run_tophat). To run the pre-installed pipelines all the way through, this should be “last_function.”
Warning
Do not change the STEPS or STEPS_DE parameters for a pre-installed pipeline.
Note
Fastq files: paired end: 2 files, “Name_1.fastq” and “Name_2.fastq” representing read 1 and read 2. Have all fastq files in same raw data folder
Warning
Default parameters have been included for each third party software tool included in each of the pipelines. Before running, please view the documentation for each software tool to determine if these parameters are appropriate for your analysis. We do not advise using the default parameters included in Omics Pipe without a full understanding of the tools/parameters.
Example Omics Pipe Parameter File¶
test_params.yaml in omics_pipe/tests:
SAMPLE_LIST: [test1, test2, test3]
STEP: last_function
STEPS: [fastqc, star, htseq, last_function]
RAW_DATA_DIR: /gpfs/home/kfisch/scripts/omics_pipeline-devel/tests
FLAG_PATH: /gpfs/home/kfisch/scripts/omics_pipeline-devel/tests/test_run/flags
HTSEQ_RESULTS: /gpfs/home/kfisch/scripts/omics_pipeline-devel/tests/test_run/counts
LOG_PATH: /gpfs/home/kfisch/scripts/omics_pipeline-devel/tests/test_run/logs
QC_PATH: /gpfs/home/kfisch/scripts/omics_pipeline-devel/tests/test_run
RESULTS_PATH: /gpfs/home/kfisch/test
STAR_RESULTS: /gpfs/home/kfisch/scripts/omics_pipeline-devel/tests/test_run/star
WORKING_DIR: /gpfs/home/kfisch/scripts/omics_pipeline-devel/omics_pipe/scripts
REPORT_RESULTS: /gpfs/home/kfisch/scripts/omics_pipeline-devel/tests/test_run
ENDS: SE
FASTQC_VERSION: '0.10.1'
GENOME: /gpfs/group/databases/Homo_sapiens/UCSC/hg19/Sequence/WholeGenomeFasta/genome.fa
HTSEQ_OPTIONS: -m intersection-nonempty -s no -t exon
PIPE_MULTIPROCESS: 100
PIPE_REBUILD: 'True'
PIPE_VERBOSE: 5
REF_GENES: /gpfs/group/databases/Homo_sapiens/UCSC/hg19/Annotation/Genes/genes.gtf
RESULTS_EMAIL: kfisch@scripps.edu
STAR_INDEX: /gpfs/group/databases/Homo_sapiens/UCSC/hg19/star_genome
STAR_OPTIONS: --readFilesCommand cat --runThreadN 8 --outSAMstrandField intronMotif --outFilterIntronMotifs RemoveNoncanonical
STAR_VERSION: '2.3.0'
TEMP_DIR: /scratch/kfisch
QUEUE: workq
USERNAME: kfisch
DRMAA_PATH: /opt/applications/pbs-drmaa/current/gnu/lib/libdrmaa.so
DPS_VERSION: '1.3.1111'
BAM_FILE_NAME: Aligned.out.bam
PARAMS_FILE: '/gpfs/home/kfisch/scripts/omics_pipeline-devel/tests/test_params_RNAseq_counts.yaml'
DESEQ_META: /gpfs/home/kfisch/scripts/omics_pipeline-devel/tests/counts_meta.csv
DESIGN: '~ condition'
PVAL: '0.05'
DESEQ_RESULTS: /gpfs/home/kfisch/scripts/omics_pipeline-devel/tests/test_run/DESEQ
SUMATRA_DB_PATH: /gpfs/home/kfisch/sumatra
SUMATRA_RUN_NAME: test_counts_sumatra_project
REPOSITORY: https://kfisch@bitbucket.org/sulab/omics_pipe
HG_USERNAME: Kathleen Fisch <kfisch@scripps.edu>
Explanation of Variables in Omics Pipe Parameter File¶
Parameters vary by pipeline and the correct parameter file for each pipeline must be used. See examples in the /tests/ folder.
RNAseq Count Based Pipeline¶
test_params_RNAseq_counts.yaml in omics_pipe/tests:
#sample names ie “Name” for paired and single end reads. So, “Name” for paired-end would expect two fastq files named “Name_1.fastq” and Name_2.fastq”
SAMPLE_LIST: [test1, test2, test3]
#Function to be run within pipeline. If you want to run the whole pipeline, leave this as last_function
STEP: last_function
#All steps within the pipeline. DO NOT CHANGE this parameter for pre-installed pipelines. If you create your own pipeline, you will need to modify this by listing all of the steps in your pipeline.
STEPS: [fastqc, star, htseq, last_function]
#Directory where your raw .fastq files are located.
RAW_DATA_DIR: /gpfs/home/kfisch/scripts/omics_pipeline-devel/tests
#Directory where you would like to have the flag files created. Flag files are empty files that indicate if a step in the pipeline has completed successfully.
FLAG_PATH: /gpfs/home/kfisch/scripts/omics_pipeline-devel/tests/test_run/flags
#Directory for HTSEQ results
HTSEQ_RESULTS: /gpfs/home/kfisch/scripts/omics_pipeline-devel/tests/test_run/counts
#Directory where log files will be written
LOG_PATH: /gpfs/home/kfisch/scripts/omics_pipeline-devel/tests/test_run/logs
#Directory for QC results
QC_PATH: /gpfs/home/kfisch/scripts/omics_pipeline-devel/tests/test_run
#Upper level results directory. Sumatra will check all subfolders of this directory for new files to add to the run tracking database.
RESULTS_PATH: /gpfs/home/kfisch/test
#Directory where STAR results will be written
STAR_RESULTS: /gpfs/home/kfisch/scripts/omics_pipeline-devel/tests/test_run/star
#Where omics_pipe is installed, this path will be pointing to ~/omics_pipe/scripts.
WORKING_DIR: /gpfs/home/kfisch/scripts/omics_pipeline-devel/omics_pipe/scripts
#Directory for the summary report
REPORT_RESULTS: /gpfs/home/kfisch/scripts/omics_pipeline-devel/tests/test_run
#SE is single end, PE is paired-end sequencing reads
ENDS: SE
#Version number of FASTQC
FASTQC_VERSION: '0.10.1'
#Full path to Genome fasta file
GENOME: /gpfs/group/databases/Homo_sapiens/UCSC/hg19/Sequence/WholeGenomeFasta/genome.fa
#Options for HTSEQ. Please see HTSEQ-count documentation for parameter options. http://www-huber.embl.de/users/anders/HTSeq/doc/count.html#options
HTSEQ_OPTIONS: -m intersection-nonempty -s no -t exon
#Number of multiple processes you want Ruffus to spawn at once
PIPE_MULTIPROCESS: 100
#Ruffus parameter. No need to change.
PIPE_REBUILD: 'True'
#Ruffus parameter. No need to change.
PIPE_VERBOSE: 5
#Full path to reference gene annotations
REF_GENES: /gpfs/group/databases/Homo_sapiens/UCSC/hg19/Annotation/Genes/genes.gtf
#Your email.
RESULTS_EMAIL: kfisch@scripps.edu
#Directory pointing to STAR_INDEX (you may have to create this)
STAR_INDEX: /gpfs/group/databases/Homo_sapiens/UCSC/hg19/star_genome
#Options for STAR. Please read parameter options https://code.google.com/p/rna-star/
STAR_OPTIONS: --readFilesCommand cat --runThreadN 8 --outSAMstrandField intronMotif --outFilterIntronMotifs RemoveNoncanonical
#Version number of STAR
STAR_VERSION: '2.3.0'
#Path to temporary directory
TEMP_DIR: /scratch/kfisch
#Name of the queue on your local cluster you wish to use
QUEUE: workq
#Username for local cluster
USERNAME: kfisch
#Path to your local cluster installation of DRMAA (ask your sys admin for this)
DRMAA_PATH: /opt/applications/pbs-drmaa/current/gnu/lib/libdrmaa.so
#Name of created Bam file. Will be Aligned.out.bam if you are using STAR and accepted_hits.bam if you are using TopHat
BAM_FILE_NAME: Aligned.out.bam
#Full path to your parameter file. Make sure to include the single quotes.
PARAMS_FILE: '/gpfs/home/kfisch/scripts/omics_pipeline-devel/tests/test_params_RNAseq_counts.yaml'
#Location of the meta data csv file for DESEQ. See tests/counts_meta.csv for an example. This file contains the Design file for your study. http://www.bioconductor.org/packages/release/bioc/html/DESeq2.html
DESEQ_META: /gpfs/home/kfisch/scripts/omics_pipeline-devel/tests/counts_meta.csv
#Design for DESEQ differential expression. Leave as is if you use the exact design as in the counts_meta.csv file.
DESIGN: '~ condition'
#P-value threshold
PVAL: '0.05'
#Directory for DESEQ results
DESEQ_RESULTS: /gpfs/home/kfisch/scripts/omics_pipeline-devel/tests/test_run/DESEQ
#Directory where you want to store your Sumatra database. Once you run this once, you do not have to change this.
SUMATRA_DB_PATH: /gpfs/home/kfisch/sumatra
#Name of your project. You do not need to change this for subsequent runs of the pipeline, but you can if you wish.
SUMATRA_RUN_NAME: test_counts_sumatra_project
#Location of omics pipe repository (you can leave this)
REPOSITORY: https://kfisch@bitbucket.org/sulab/omics_pipe
#Your Mercurial username
HG_USERNAME: Kathleen Fisch <kfisch@scripps.edu>
#Version of Python installed on system
PYTHON_VERSION: 2.6.5
#Type of cluster scheduler (options: PBS, SGE)
SCHEDULER: PBS
#Full path to WORKING_DIR/reporting on your system
R_SOURCE_PATH: /gpfs/home/kfisch/scripts/omics_pipeline-devel/omics_pipe/scripts/reporting
#Are your raw fastq files compressed? If not, leave this parameter as none. If so, please type 'GZIP' and it will automatically process your gzip files.
COMPRESSION: none
RNAseq Tuxedo Pipeline¶
test_params_RNAseq_Tuxedo.yaml in omics_pipe/tests:
#sample names ie “Name” for paired and single end reads. So, “Name” for paired-end would expect two fastq files named “Name_1.fastq” and Name_2.fastq”
SAMPLE_LIST: [test1, test2]
#Function to be run within pipeline. If you want to run the whole pipeline, leave this as last_function
STEP: last_function
#All steps within the pipeline. DO NOT CHANGE this parameter for pre-installed pipelines. If you create your own pipeline, you will need to modify this by listing all of the steps in your pipeline.
STEPS: [fastqc, tophat, rseqc, cufflinks]
#All steps within the pipeline. DO NOT CHANGE this parameter for pre-installed pipelines. If you create your own pipeline, you will need to modify this by listing all of the steps in your pipeline.
STEPS_DE: [cuffmerge, cuffmergetocompare, cuffdiff, RNAseq_report_tuxedo, last_function]
#SE is single end, PE is paired-end sequencing reads
ENDS: SE
#Your email address
RESULTS_EMAIL: kfisch@scripps.edu
#Path to temporary directory (make sure this is a large, writable directory)
TEMP_DIR: /scratch/kfisch
#Name of the queue on your cluster
QUEUE: workq
#Your username on the cluster
USERNAME: kfisch
#Full path to your raw data files
RAW_DATA_DIR: /gpfs/home/kfisch/scripts/omics_pipeline-devel/tests
#Full path to where you want the Flag files written
FLAG_PATH: /gpfs/home/kfisch/scripts/omics_pipeline-devel/tests/test_run
#Full path to where you want the log files for each step written
LOG_PATH: /gpfs/home/kfisch/scripts/omics_pipeline-devel/tests/test_run
#Full path to where you want QC results written
QC_PATH: /gpfs/home/kfisch/scripts/omics_pipeline-devel/tests/test_run
#Full path to upper level results path
RESULTS_PATH: /gpfs/home/kfisch/scripts/omics_pipeline-devel/tests/test_run
#Where omics_pipe is installed, this path will be pointing to ~/omics_pipe/scripts
WORKING_DIR: /gpfs/home/kfisch/scripts/omics_pipeline-devel/omics_pipe/scripts
#Full path to where you want Tophat results written
TOPHAT_RESULTS: /gpfs/home/kfisch/scripts/omics_pipeline-devel/tests/test_run/alignments
#Full path to where you want Cufflinks results written
CUFFLINKS_RESULTS: /gpfs/home/kfisch/scripts/omics_pipeline-devel/tests/test_run/assemblies
#Full path to where you want Cuffmerge results written
CUFFMERGE_RESULTS: /gpfs/home/kfisch/scripts/omics_pipeline-devel/tests/test_run/cuffmerge
#Full path to where you want Cuffdiff results written
CUFFDIFF_RESULTS: /gpfs/home/kfisch/scripts/omics_pipeline-devel/tests/test_run/cuffdiff
#List of full paths to alignment files divided by condition. Each sample will have the path TOPHAT_RESULTS/SAMPLE_NAME/accepted_hits.bam. Divide these up into your two conditions for differential expression analysis. See http://cole-trapnell-lab.github.io/cufflinks/cuffdiff/#cuffdiff-arguments for more details.
CUFFDIFF_INPUT_LIST_COND1: /gpfs/home/kfisch/scripts/omics_pipeline-devel/tests/test_run/alignments/test1/accepted_hits.bam
CUFFDIFF_INPUT_LIST_COND2: /gpfs/home/kfisch/scripts/omics_pipeline-devel/tests/test_run/alignments/test2/accepted_hits.bam
#Options for Tophat. Please read the TopHat documentation to set these options for your analysis. http://ccb.jhu.edu/software/tophat/manual.shtml#toph
TOPHAT_OPTIONS: -p 8 -a 5 --microexon-search --library-type fr-secondstrand
CUFFLINKS_OPTIONS: -u -N
CUFFMERGE_OPTIONS: -p 8
CUFFMERGETOCOMPARE_OPTIONS: -CG
CUFFDIFF_OPTIONS: -p 8 -FDR 0.01 -L Group1,Group2 -N --compatible-hits-norm
#Software versions installed on your system
FASTQC_VERSION: '0.10.1'
TOPHAT_VERSION: '2.0.9'
CUFFLINKS_VERSION: '2.1.1'
R_VERSION: '3.0.1'
BOWTIE_VERSION: 2.2.3
SAMTOOLS_VERSION: 0.1.19
PYTHON_VERSION: 2.6.5
BOWTIE_VERSION: 2.2.3
SAMTOOLS_VERSION: 0.1.19
#Ruffus specific parameters. See above or documentation for details. http://www.ruffus.org.uk/pipeline_functions.html#pipeline-functions-pipeline-run
PIPE_MULTIPROCESS: 100
PIPE_REBUILD: 'True'
PIPE_VERBOSE: 5
#Full path to gene annotation gtf file
REF_GENES: /gpfs/group/databases/Homo_sapiens/UCSC/hg19/Annotation/Genes/genes.gtf
#Full path to genome file
GENOME: /gpfs/group/databases/Homo_sapiens/UCSC/hg19/Sequence/WholeGenomeFasta/genome.fa
#Full path to BOWTIE index
BOWTIE_INDEX: /gpfs/group/databases/Homo_sapiens/UCSC/hg19/Sequence/Bowtie2Index/genome
#Location of chromosomes folder
CHROM: /gpfs/group/databases/Homo_sapiens/UCSC/hg19/Sequence/Chromosomes
#Path to your local cluster installation of DRMAA (ask your sys admin for this)
DRMAA_PATH: /opt/applications/pbs-drmaa/current/gnu/lib/libdrmaa.so
#Full path to directory where you want report results written
REPORT_RESULTS: /gpfs/home/kfisch/scripts/omics_pipeline-devel/tests
#Full path to your parameter file. Make sure to include the single quotes.
PARAMS_FILE: '/gpfs/home/kfisch/scripts/omics_pipeline-devel/tests/test_params_RNAseq_Tuxedo.yaml'
#Gene IDs of interest for visualization with CummeRbund
GENEIDS: [GAPDH, COL2A1, BRCA2]
#Name of samples in Condition 1
COND1: Group1
#Name of samples in Condition 2
COND2: Group2
#Name of your project. You do not need to change this for subsequent runs of the pipeline, but you can if you wish.
NAME: Test_run_date
#Directory where you want to store your Sumatra database. Once you run this once, you do not have to change this.
SUMATRA_DB_PATH: /gpfs/home/kfisch/sumatra
#Name of your project. You do not need to change this for subsequent runs of the pipeline, but you can if you wish.
SUMATRA_RUN_NAME: default_sumatra_project
#Location of omics pipe repository (you can leave this)
REPOSITORY: https://kfisch@bitbucket.org/sulab/omics_pipe
#Your Mercurial username
HG_USERNAME: Kathleen Fisch <kfisch@scripps.edu>
#Type of cluster scheduler (options: PBS, SGE)
SCHEDULER: PBS
#Full path to WORKING_DIR/reporting on your system
R_SOURCE_PATH: /gpfs/home/kfisch/scripts/omics_pipeline-devel/omics_pipe/scripts/reporting
#Are your raw fastq files compressed? If not, leave this parameter as none. If so, please type 'GZIP' and it will automatically process your gzip files.
COMPRESSION: none
