Tutorial: GotCloud

From Genome Analysis Wiki
Revision as of 00:58, 8 January 2013 by Amkwong (talk | contribs)
Jump to navigationJump to search
The printable version is no longer supported and may have rendering errors. Please update your browser bookmarks and please use the default browser print function instead.

Installation

First, make sure GotCloud is installed on your system. Installation instructions here.


Running the Automatic Test

This verifies that GotCloud was installed correctly.

To run the test case automatically, change your current directory to GotCloud's root directory, and type in the following command:

/bin/gen_biopipeline.pl --test OUTPUT_DIR

where OUTPUT_DIR is the directory where you want to store the results.

If you see "Test Passed", then you are ready to run a sample.


Aligning a Sample

As an example, we can analyze the sample files used in the automatic test.

To make this easier, change to the {ROOT_DIR}/test/align directory. (We will call the directory in which GotCloud is installed "{ROOT_DIR}".) It contains an index file and a configuration file that can be used directly.

Index file

There are four fastq files in {ROOT_DIR}/test/align/fastq/Sample_1 and four fastq files in {ROOT_DIR}/test/align/fastq/Sample_2, both in paired-end format. Normally, we would need to build an index file for these files. Conveniently, an index file (indexFile.txt) already exists for the automatic test samples. It can be found in {ROOT_DIR}/test/align/, and contains the following information in tab-delimited format:

MERGE_NAME FASTQ1                           FASTQ2                           RGID   SAMPLE    LIBRARY CENTER PLATFORM
Sample1    fastq/Sample_1/File1_R1.fastq.gz fastq/Sample_1/File1_R2.fastq.gz RGID1  SampleID1 Lib1    UM     ILLUMINA
Sample1    fastq/Sample_1/File2_R1.fastq.gz fastq/Sample_1/File2_R2.fastq.gz RGID1a SampleID1 Lib1    UM     ILLUMINA
Sample2    fastq/Sample_2/File1_R1.fastq.gz fastq/Sample_2/File1_R2.fastq.gz RGID2  SampleID2 Lib2    UM     ILLUMINA
Sample2    fastq/Sample_2/File2_R1.fastq.gz fastq/Sample_2/File2_R2.fastq.gz RGID2  SampleID2 Lib2    UM     ILLUMINA

If you are in the {ROOT_DIR}/test/align directory, you can use this file as-is. If you prefer, you can create a new index file and change the MERGE_NAME, RGID, SAMPLE, LIBRARY, CENTER, or PLATFORM values. It is recommended that you do not modify existing files in {ROOT_DIR}/test/align.

If you want to run this example from a different directory, make sure the FASTQ1 and FASTQ2 paths are correct. That is, each of the FASTQ1 and FASTQ2 entry in the index file should look like the following:

{ROOT_DIR}/test/align/fastq/Sample_1/File1_R1.fastq.gz 

Alternately, if you want to run this example from a different directory, but do not want to edit the index file, you can copy all the fastq files to a new directory with the relative path listed in the index file:

ln -s {ROOT_DIR}/test/align/fastq fastq

This will create a symbolic link to the test fastq directory from your current directory.

Configuration file

Similar to the index file, a configuration file (test.conf) already exists for the automatic test samples. It contains the following information:

INDEX_FILE = indexFile.txt
############
# References
REF_DIR = $(PIPELINE_DIR)/test/align/chr20Ref
AS = NCBI37
FA_REF = $(REF_DIR)/human_g1k_v37_chr20.fa
DBSNP_VCF =  $(REF_DIR)/dbsnp.b130.ncbi37.chr20.vcf.gz
PLINK = $(REF_DIR)/hapmap_3.3.b37.chr20

If you are in the test/align directory, you can use this file as-is. If you are using a different index file, make sure your index file is named correctly in the first line.

Running the alignment pipeline

You are now ready to run the alignment pipeline. This requires two steps: first, generating the Makefiles; and second, running those Makefiles.

Generating the Makefiles

Enter the following command:

{ROOT_DIR}/bin/gen_biopipeline.pl --conf test.conf --out_dir {OUT_DIR}

where {OUT_DIR} is the directory in which you wish to store the resulting BAM files.

If everything went well, you will see the following messages:

Finished creating makefile {OUTDIR}/Makefiles/biopipe_Sample2.Makefile
Finished creating makefile {OUTDIR}/Makefiles/biopipe_Sample1.Makefile
--------------------------------------------------------------------
Run the following commands:

make -f {OUTDIR}/Makefiles/biopipe_Sample2.Makefile > {OUTDIR}/Makefiles/biopipe_Sample2.Makefile.log
make -f {OUTDIR}/Makefiles/biopipe_Sample1.Makefile > {OUTDIR}/Makefiles/biopipe_Sample1.Makefile.log

where {OUTDIR} will be replaced with the directory you entered above.

Running the Makefiles

To run a Makefile, simply enter one-by-one the commands generated in the previous step. The log files for the runs will be found in the Makefiles directory, while the BAM files will be found in the {OUT_DIR}/alignment.recal directory.


Analyzing a Sample

Using umake, you can analyze the BAM files generated in the previous step and generate a VCF file.

Index file