Pipeline for mouse genotyping assay

Analyses implemented in the following study:

Andrews KR, Hunter SS, Torrevillas BK, Cespedes N, Garrison SM, Strickland J, Wagers D, Hansten G, New DD, Fagnan MW, Luckhart S. A new mouse SNP genotyping assay for speed congenics: combining flexibility, affordability, and power. In review.

Requirements

Programs

• Python v.2.7
• Fastqc
• MultiQC
• HTStream v.1.1.0
• BWA v.0.7.17
• Samtools v.1.5
• GATK v.4.1.3.0
• R v.3.6.0
• R packages:
  • gdsfmt
  • SNPRelate
  • pheatmap
  • RColorBrewer
  • plyr
  • tidyverse

Scripts

• 01-cleaning.py
• 02-map_reads.py
• 03.0-gatk.py
• 03.1-combine-gatk.py
• 03.2-call-variants.py
• 04-analyze_VCF.R
• 05-GenotypeSummary.R

Other Files

• Raw fastq.gz files, demultiplexed by sample and stored in ./00-RawData
• Reference genome (indexed with BWA), stored as ./ref/reference.fasta
• Bed file of SNP targets: ST2181G_1_target_ST2181G_1_2053-noextend.bed
• Illumina adapter sequences: adapters.fa

Genotyping Pipeline

Quality assessment

mkdir -p 00-fastqc
fastqc -o ./00-fastqc ./00-RawData/*
multiqc -i fastqc ./00-fastqc/

Cleaning

python 01-cleaning.py
bash 01-cleaning_commands.sh

Mapping

python 02-map_reads.py
bash 02-mapping_commands.sh

Quality assessment of mapping

multiqc -d ./02-mapped/ -i Mapping -o ./02-mapped/

Build samtools stats for all bam files

for f in ./02-mapped/*.bam
do
samtools stats $f > $f.stats &
done

Index all bam files

cd ./02-mapped
for f in *.bam
do
samtools index $f &
done
cd ../

Genotype with GATK

python 03.0-gatk.py
bash 03.0-gatk_command.sh

python 03.1-combine-gatk.py
bash 03.1-gatk_combine.sh

python 03.2-call-variants.py
bash 03.2-gatk_call.sh

Calculate depth by SNP

samtools depth -a -b ST2181G_1_target_ST2181G_1_2053-noextend.bed ./02-mapped/*.bam > target_depth.tsv
ls -la ./02-mapped/*.bam > samples_bam.txt

Analyze VCF

mkdir -p 04-deliverables
Rscript 04-analyze_VCF.R

Outputs:

• Genotypes for each sample and SNP: genotype.tsv
• Plots of coverage depth per SNP per sample: coverage_assessment.pdf
• Plot of genotypes for each sample and SNP: genotype_plots-ordered.png
• List of poorly performing SNPs (SNPs with coverage < 5 in more than 30 samples): poor_targets.tsv

Rscript 05-GenotypeSummary.R

Outputs:

• Ancestry proportions for each sample: Genotype_Summary_Table.tsv