Overview¶

GBRS is a suite of tools for reconstructing genomes using RNA-Seq data from multiparent population and quantifying allele specific expression. Although we tested it with mouse models only, GBRS should work for any multiparent populations. If you are interested in using it for other multiparent population model, please contact Kwangbom “KB” Choi <kb.choi@jax.org> at The Jackson Laboratory. For the Diversity Outbred (DO https://www.jax.org/strain/009376), Collaborative Cross (CC) mice, or F1 hybrids of CC’s (CCRIX), required data files are available at ftp://churchill-lab.jax.org/pub/software/GBRS/.

• Free software: GPLv3 license
• Documentation: https://gbrs.readthedocs.org.

References¶

Manuscript in preparation (Expected in 2016)

To use GBRS in command line¶

Note: We will assume you installed GBRS in its own conda virtual environment. First of all, you have to “activate” the virtual environment by doing the following:

source activate gbrs

The first step is to align our RNA-Seq reads against pooled transcriptome of all founder strains:

bowtie -q -a --best --strata --sam -v 3 ${GBRS_DIR}/bowtie.transcriptome ${FASTQ} \
    | samtools view -bS - > ${BAM_FILE}

Before quantifying multiway allele specificity, bam file should be converted into emase format:

gbrs bam2emase -i ${BAM_FILE} \
               -m ${GBRS_DATA}/ref.transcripts.info \
               -s ${COMMA_SEPARATED_LIST_OF_HAPLOTYPE_CODES} \
               -o ${EMASE_FILE}

We can compress EMASE format alignment incidence matrix:

gbrs compress -i ${EMASE_FILE} \
              -o ${COMPRESSED_EMASE_FILE}

If storage space is tight, you may want to delete ${BAM_FILE} or ${EMASE_FILE} at this point since ${COMPRESSED_EMASE_FILE} has all the information the following steps would need. If you want to merge emase format files in order to, for example, pool technical replicates, you run ‘compress’ once more listing files you want to merge with commas:

gbrs compress -i ${COMPRESSED_EMASE_FILE1},${COMPRESSED_EMASE_FILE2},... \
              -o ${MERGED_COMPRESSED_EMASE_FILE}

and use ${MERGED_COMPRESSED_EMASE_FILE} in the following steps. Now we are ready to quantify multiway allele specificity:

gbrs quantify -i ${COMPRESSED_EMASE_FILE} \
              -g ${GBRS_DATA}/ref.gene2transcripts.tsv \
              -L ${GBRS_DATA}/gbrs.hybridized.targets.info \
              -M 4 --report-alignment-counts

Then, we reconstruct the genome based upon gene-level TPM quantities (assuming the sample is a female from the 20th generation Diversity Outbred mice population)

gbrs reconstruct -e gbrs.quantified.multiway.genes.tpm \
                 -t ${GBRS_DATA}/tranprob.DO.G20.F.npz \
                 -x ${GBRS_DATA}/avecs.npz \
                 -g ${GBRS_DATA}/ref.gene_ids.ordered.npz

We can now quantify allele-specific expressions on diploid transcriptome:

gbrs quantify -i ${COMPRESSED_EMASE_FILE} \
              -G gbrs.reconstructed.genotypes.tsv \
              -g ${GBRS_DATA}/ref.gene2transcripts.tsv \
              -L ${GBRS_DATA}/gbrs.hybridized.targets.info \
              -M 4 --report-alignment-counts

Genotype probabilities are on a grid of genes. For eQTL mapping or plotting genome reconstruction, we may want to interpolate probability on a decently-spaced grid of the reference genome.:

gbrs interpolate -i gbrs.reconstructed.genoprobs.npz \
                 -g ${GBRS_DATA}/ref.genome_grid.64k.txt \
                 -p ${GBRS_DATA}/ref.gene_pos.ordered.npz \
                 -o gbrs.interpolated.genoprobs.npz

To plot a reconstructed genome:

gbrs plot -i gbrs.interpolated.genoprobs.npz \
          -o gbrs.plotted.genome.pdf \
          -n ${SAMPLE_ID}

To use GBRS in a project¶

All the features are available as a python module. You can simply:

import gbrs

Types of Contributions¶

Get Started!¶

Ready to contribute? Here’s how to set up gbrs for local development.

1. Fork the gbrs repo on GitHub.

2. Clone your fork locally:

   $ git clone git@github.com:your_name_here/gbrs.git
   

3. Install your local copy into a virtualenv. Assuming you have virtualenvwrapper installed, this is how you set up your fork for local development:

   $ mkvirtualenv gbrs
   $ cd gbrs/
   $ python setup.py develop
   

4. Create a branch for local development:

   $ git checkout -b name-of-your-bugfix-or-feature
   

   Now you can make your changes locally.

5. When you’re done making changes, check that your changes pass flake8 and the tests, including testing other Python versions with tox:

   $ flake8 gbrs tests
   $ python setup.py test
   $ tox
   

   To get flake8 and tox, just pip install them into your virtualenv.

6. Commit your changes and push your branch to GitHub:

   $ git add .
   $ git commit -m "Your detailed description of your changes."
   $ git push origin name-of-your-bugfix-or-feature
   

7. Submit a pull request through the GitHub website.

Pull Request Guidelines¶

Before you submit a pull request, check that it meets these guidelines:

1. The pull request should include tests.
2. If the pull request adds functionality, the docs should be updated. Put your new functionality into a function with a docstring, and add the feature to the list in README.rst.
3. The pull request should work for Python 2.6, 2.7, 3.3, 3.4 and 3.5, and for PyPy. Check https://travis-ci.org/kbchoi-jax/gbrs/pull_requests and make sure that the tests pass for all supported Python versions.

Tips¶

To run a subset of tests:

$ python -m unittest tests.test_gbrs

Development Lead¶

• Kwangbom “KB” Choi, Ph. D. <kb.choi@jax.org>, The Jackson Laboratory

Contributors¶

None yet. Why not be the first?

0.1.5 (05/04/2016)¶

• compress subcommand also merges emase files
• reconstruct does not require ref.gene_ids.ordered.npz file anymore

0.1.4 (05/04/2016)¶

• Uploaded to Anaconda Cloud

0.1.2 (04/27/2016)¶

• Setup Travis CI automated building

0.1.1 (04/27/2016)¶

• Moved to github churchill-lab space
• First release onto PyPI

0.1.0 (01/22/2016)¶

• Started a private project on bitbucket.