An R-based bioinformatic pipeline to determine killer-cell immunoglobulin-like receptor (KIR) copy number and high-resolution genotypes from short-read sequencing data.
Paired-end KIR targeted sequencing data
Download the pipeline code with the following line:
git clone https://github.com/Hollenbach-lab/PING.git
cd ./PINGTo ensure a reliable run, we have containerized our image in Singularity (tested on version 3.11.4). You can check whether Singularity has been installed in your system by running singularity --version. Please install Singularity by following their guide here.
Once Singularity has been installed, obtain the image using one of these following commands:
--fakeroot
singularity build --fakeroot ping.sif ping.defsingularity pull ping.sif library://rsuseno/rsuseno/ping:latestsudo singularity build ping.sif ping.defEnsure that you are within the PING directory and you can run the entirety of the pipeline using the following command:
singularity exec --bind <fastq_location> ping.sif Rscript PING_run.R
--fqDirectory <fastq_location>
--resultsDirectory <output_location>
--fastqPattern <fastq_pattern>
--threads <number_of_threads>Listed below are the arguments needed to run PING:
--fqDirectory Set to raw sequence directory or extracted fastq directory if extraction has already been performed--resultsDirectory Set the results directory, one will be created if it does not already exist (all pipeline output will be recorded here)--fastqPattern Specify a pattern on the fqDirectory to only process specific samples. For example, if your sequencing data is named [SAMPLE_ID]_R1_fq.gz, you would change it to 'fq'. Additionally, you can use 'KIR' to find already extracted files. (default = 'fastq')--threads Number of threads to use during bowtie2 alignments (default = 4)--bind Needed when the input FASTQs are located outside of your access privilege (e.g., if the inputs are not within /home/username). It is recommended to pass the --bind option regardless to ensure a proper run.We have included 10 test sequences to run through the pipeline, they are located in the test_sequence/ directory. These samples are meant to test that all the installations were done properly. You can run the following code to execute the test:
singularity exec ping.sif Rscript PING_run.R
--fqDirectory test_sequence
--resultsDirectory test_sequence_output Copy number output can be found at [resultsDirectory]/predictedCopyNumberFrame.csv
Genotype output can be found at [resultsDirectory]/finalAlleleCalls.csv
Aligned SNP tables can be found in [resultsDirectory]/alignmentFiles/[sampleID]/iterAlign/
Copy number graphs can be found in [resultsDirectory]/copyPlots/
If PING is unable to perfectly match aligned SNPs to known KIR allele sequences an unresolved call will be produced.
Unresolved genotype information can be found in [resultsDirectory]/iterAlleleCalls.csv, where the closest allele match is recorded along with the mismatched SNP information in the following format:
[closest_matched_allele]$[exon]_[position].[nucleotide]
Where closest matched allele is the allele genotyping that best matches the aligned SNPs, nucleotide denotes the mismatched nucleotide located at the indicated exon and position within the exon. Multiple mismatched SNPs are connected with the ^ symbol.
Please save a copy of your terminal output and contact me through Github or email at wesley.marin@ucsf.edu or rayo.suseno@ucsf.edu.
Please cite:
Marin WM, Dandekar R, Augusto DG, Yusufali T, Heyn B, et al. (2021) High-throughput Interpretation of Killer-cell Immunoglobulin-like Receptor Short-read Sequencing Data with PING. PLOS Computational Biology 17(8): e1008904. https://doi.org/10.1371/journal.pcbi.1008904
Robinson J, Waller MJ, Stoehr P, Marsh SGE. IPD-the Immuno Polymorphism Database. Nucleic Acids Research (2005), 331:D523-526
Langmead B, Salzberg S. Fast gapped-read alignment with Bowtie 2. Nature Methods. 2012, 9:357-359.