This is a project for a pipeline of high-resolution KIR allele annotations on human genome assemblies.

It uses several libraries including argparse, pandas, re, gzip, and Biopython. The script provides options for outputting in different formats, including BED, VCF, and HAP(CSV).

• BioPython
• pandas
• argparse

You can install these packages using pip:

pip install biopython pandas argparse

git clone https://github.com/calvinckhung/skirt.git

Export your SKIRT directory

echo "export SKIRT_WD=<Your SKIRT directory>"

We already include the 2.12.0 IPDKIR in the project. In case you want to update from scratch, clone the IPDKIR repository to the SKIRT directory:

cd skirt
git clone https://github.com/ANHIG/IPDKIR.git

This will overwrite the included IPDKIR files. Be careful.

For installation of minimap2, refer to its GitHub page https://github.com/lh3/minimap2. Ensure your $PATH variable contains the path to the minimap2 executable binary. Otherwise, export this variable:

echo "export MM2_BIN_PATH=<Your minimap2 binary path>"
echo "export PATH=$PATH:$MM2_BIN_PATH" >> ~/.bashrc

Don't forget to install tblastn, part of the BLAST+ suite:

cd ~/bin
wget ftp://ftp.ncbi.nlm.nih.gov/blast/executables/blast+/2.14.0/ncbi-blast-2.14.0+-x64-linux.tar.gz
tar -xzf ncbi-blast-2.14.0+-x64-linux.tar.gz
echo "export BLAST_BIN_PATH=<Path to tblastn executable binary>"
echo "export PATH=$PATH:$BLAST_BIN_PATH" >> ~/.bashrc

For a detailed installation description of BLAST+, please refer to https://www.ncbi.nlm.nih.gov/books/NBK569861/ If you have BLAST+ installed already, don't forget to specify your path as above.

To run SKIRT KIR allele annotation for your own assembly sequence, run below command under SKIRT directory:

./scripts/miniskirt.sh <path to the assembly sequence (.fa or .fa.gz)> <output path>

You will get *.BED, *.VCF and *.Hap.CSV files. Usually takes between 5~20 minutes.