Skip to content

Latest commit

 

History

History
93 lines (67 loc) · 3.79 KB

hla_typing.md

File metadata and controls

93 lines (67 loc) · 3.79 KB

HLA Typing exercise using Polysolver and Optitype

The samples can be found at /etc/ace-data/CancerGenomicsWG/VariantCalling/samples/Validation. For demonstration, I will call my sample as sample, and its corresponding reads as sample_R1.fastq and sample_R2.fastq for forward and reverse reads respectively. Modify this according to your assigned sample.

PolySolver

Steps 1: Alignment with BWA:

We first align our samples to the human reference genome, version 19 (hg19). A copy of this reference and its indices have been added to our shared folder under the location: /etc/ace-data/CancerGenomicsWG/Tools/References. However, feel free to use your own copy of the reference in case you have one.

# Load the bwa and samtools modules
module load bwa-0.7.17-gcc-8.5.0-h3gmzcf samtools-1.13-gcc-8.5.0-hx66cfb

mkdir -p Results/Alignment_hg19

bwa mem -t 16 /etc/ace-data/CancerGenomicsWG/Tools/References/hg19.fa  sample_R1.fastq sample_R2.fastq  -o Results/Alignment_hg19/sample.sam
samtools sort -@ 12 Results/Alignment_hg19/sample.sam -o Results/Alignment_hg19/sample.bam
samtools index Results/Alignment_hg19/sample.bam

Step 2: Create a Polysolver conda environment

We will need to first create a conda environment for this.

conda env create --name polysolver  --file /etc/ace-data/CancerGenomicsWG/Tools/polysolver.yml --yes

Step 3: Run the Polysolver script

When the environment is done being created, go ahead and activate it and run the polysolver hla typing script.

#Activate the environmnet
#conda activate polysolver
conda activate /etc/ace-data/CancerGenomicsWG/Tools/polysolver_env

mkdir -p Results/polysolver

#Run the polysolver script
shell_call_hla_type Results/Alignment_hg19/sample.bam Black 1 hg19 STDFQ 0 Results/polysolver/sample

When the script is done running, check for the file called winners.hla.txt in the output directory specified (ie Results/polysolver/sample) in our command above. Its contents will look something like this:

HLA-A   hla_a_01_04n    hla_a_01_04n
HLA-B   hla_b_07_03     hla_b_07_03
HLA-C   hla_c_01_03     hla_c_01_03

Optitype

Read more about optitype at their Github repo

Step 1: Creating an Optitype Conda environment

conda env create --name optitype  --file /etc/ace-data/CancerGenomicsWG/Tools/optitype.yml --yes

Step 2: Running the Optitype tool

#conda activate optitype
conda activate /etc/ace-data/CancerGenomicsWG/Tools/optitype_env

python /etc/ace-data/CancerGenomicsWG/Tools/OptiType/OptiTypePipeline.py \
              -i sample_R1.fastq  sample_R2.fastq \
              --dna -v -o ./Results/Optitype -c /etc/ace-data/CancerGenomicsWG/Tools/OptiType/config.ini --prefix sample

A few things to note:

  • Clone the Github repository for optitype git clone https://github.com/FRED-2/OptiType.git
  • Install all required software and libraries including Python 2.7 after activating the conda environment (Most of these softwares and libraries can be installed using conda)
  1. RazerS 3.4
  2. SAMtools 1.2
  3. HDF5 1.8.15
  4. CPLEX 12.5 or other Pyomo-supported ILP solver (GLPK, CBC, ...
  • Install all the Python modules after activating the conda environment
  1. pip install numPy==1.9.3
  2. pip install pyomo
  3. conda install pytables
  4. pip install pandas
  5. pip install pysam==0.8.3
  6. pip install matplotlib
  7. pip install future

  • Change the directory to that folder: cd Optitype inorder to be able to use the optitype script OptiTypePipeline.py or you can as wll use the full path to the optitype script

  • Make a copy of the config.ini file and include your own version the razers3 program installed in your space, that is razers3=/path/to/razers3

  • Modify the input files and sample ids to match your sample assigned.