Modify ClinVar VCF for ingestion into TileDB-VCF

Makefile

# genome reference
fai_url := s3://broad-references/hg38/v0/Homo_sapiens_assembly38.fasta.fai
fai_file := $(notdir $(fai_url))

# original clinvar vcf
vcf_url := ftp.ncbi.nlm.nih.gov/pub/clinvar/vcf_GRCh38/clinvar_20210710.vcf.gz
raw_vcf := $(notdir $(vcf_url))

# modified vcf file
mod_vcf := $(raw_vcf:%.vcf.gz=%_modified.vcf.gz)

all: $(fai_file) $(raw_vcf) $(mod_vcf)

$(mod_vcf): temp.vcf.gz temp.vcf.gz.tbi $(fai_file)
	@echo "Fix contig lengths in $@"
	@bcftools reheader -f $(fai_file) $< -o $@
	@bcftools index --tbi $@

temp.vcf: $(raw_vcf) $(addsuffix .tbi, $(raw_vcf))
	@echo "Creating modified VCF: $@"
	@echo "..create new header with FORMAT/GT defined in $@"
	@bcftools view --header-only $< | grep "^##" > $@
	@echo "##FORMAT=<ID=GT,Number=1,Type=String,Description=\"Genotype\">" >> $@
	@echo "..add dummy sample column to $@"
	@echo "#CHROM\tPOS\tID\tREF\tALT\tQUAL\tFILTER\tINFO\tFORMAT\tsample1" >> $@
	@bcftools view --no-header $< \
		| grep -v '^NW_009646201.1' \
		| awk 'BEGIN{OFS="\t"}{print $$0, "GT", "./."}' >> $@

$(raw_vcf):
	@echo "Downloading raw VCF $@"
	@curl -o $@ $(vcf_url)

$(fai_file):
	@echo "Downloading $@"
	@aws s3 cp $(fai_url) $@
	@echo "Removing chr prefixes"
	@sed -i '' 's|^chr||g' $@
	@echo "Renaming mitochondrial chr to MT"
	@sed -i '' 's|^M|MT|g' $@

%.vcf.gz: %.vcf
	@echo "Compressing $<"
	@bcftools view -Oz -o $@ $<

%.vcf.gz.tbi: %.vcf.gz
	@echo "Creating index: $@"
	@bcftools index --tbi $<

clean:
	rm -rf $(fai_file) $(mod_vcf) $(raw_vcf) *.tbi

.PHONY: clean
.INTERMEDIATE: temp.vcf temp.vcf.gz temp.vcf.gz.tbi