Software for analysis of Droplet Barcode Sequencing data.

    git clone https://github.com/elhb/DBS_Analysis.git
    cd DBS_Analysis
    pip install -r requirements.txt
    python setup.py build
    python setup.py install

Note that: Some of the scripts make use of other non-python software you therefore need working installations of bowtie2, cd-hit-454 and picard tools (version 1.114) to be able to run all commands. It's easy to install all of this using conda, take a look at the analysis_automation.sh script. Otherwise you can find the software here:

• bowtie2 can be found here
• picard tools can be found here
• cd-hit-454 can be found here Note that: the OpenMP threading of cd-hit does not work for the conda installation.

This script is supplied to allow others to reproduce the analysis exactly as it was done for the publication.

    bash analysis_automation.sh

Note that: This part is still under heavy development.

For each sample you want to analyze the following steps need to be performed <path to analysis> is the path to a folder where you want your analysis results and intermediary files to be stored.

To review your settings or see what options to work with run the following command:

    dbs_change_settings <path to analysis> listsettings

To set a value for a variable run the following command:

    dbs_change_settings <path to analysis> <variable name>=<value>

Example: Remember that the code supplied below is only an example and the paths most probably need to be changed, eg. the file "reference_data/hla_a.fasta" is probably not around on your computer at this moment.

    dbs_change_settings ~/my_analysis/sample_one
        mapqCutOff=20 \
        maxHandleMissMatches=2 \
        maxIndividualIndexMissMatches=1 \
        barcodeMissmatch=2 \
        port=5000 \
        targetRegionBed=software/DBS_Analysis/goodies/target_regions.bed \
        IndexReferenceTsv=software/DBS_Analysis/goodies/indexRef.tsv \
        bowtie2Reference=reference_data/hla_a.fasta \
        picardPath=software/picard-tools-1.114/ \
        known_hla_types=reference_data/ipd.imgt.hla_A_gen.fasta \
        minReadDepthForVariantCalling=5 \
        minPairsPerCluster=20;

    dbs_handle_indetifier <path to read one fastq> <path to read two fastq> <path to analysis>;

Note that: If you use the keyword SKIP instead of <path to read one fastq> files will not be added to the database but the handle and dbs identification will be run for all files already in the database:

    dbs_handle_indetifier SKIP SKIP <path to analysis>;

Note that: If you use the keyword justadd at the end of the line files only be added to the database and no handle and dbs identification will be performed:

    dbs_handle_indetifier <path to read one fastq> <path to read two fastq> <path to analysis> justadd;

    dbs_barcode_clustrer <path to analysis>

    dbs_insert_mapper <path to analysis>

    dbs_cluster_analyzer <path to analysis>

Note that: It is sometimes usefull to be able to rerun this step from scratch by supplying the argument reanalyze=True:

    dbs_cluster_analyzer <path to analysis> reanalyze=True

Note that: This step sometimes freezez before finishing (eg. at 60% progress) this usually indicates that something is funky with the pysam installation, if this is the case try to build pysam from source, potentially also try switching virtualenv to conda this worked for me at several occasions.

    dbs_find_alleles <path to analysis>

Note that: You can also supply the optional keyword graph to skip some of the computiationally intenst steps and just make some plots:

    dbs_find_alleles <path to analysis> [graph]

    dbs_hla_server <path to analysis>

    dbs_downsample_bam <path to analysis>

this adds a barcode cluster specififc tag ("bc") to each read in the <path to analysis>/data/mappedInserts.bam creating a new file named <path to analysis>/data/mappedInserts.tagged.bam

    dbs_tag_bam <path to analysis>

    viewLogfiles <path to analysis>/logfiles