Bcftools Variant Manipulation

# Bioconda (recommended — installs HTSlib suite)
conda install -c bioconda bcftools

# Homebrew (macOS)
brew install bcftools

# Verify
bcftools --version | head -1
# bcftools 1.20

# Index a VCF for region queries
bcftools index -t variants.vcf.gz   # creates .tbi
bcftools index -c variants.vcf.gz   # creates .csi (for chromosomes > 512 Mb)

decisions:
  - id: D1
    param: filterExpression
    kind: required
    source: user
    ask: "Which variants should survive - by depth, genotype quality, allele balance, or population frequency?"
    default: null

  - id: D2
    param: multiallelicHandling
    kind: required
    source: user
    ask: "Should multi-allelic records be split into one row per alternate allele before filtering and annotation?"
    default: "split - most annotation and comparison tools assume one alternate per row"

  - id: D3
    param: leftAlignment
    kind: required
    source: upstream
    depends_on: [D2]
    ask: "Should indels be left-aligned against the reference so the same variant is written identically across callers?"
    default: "left-aligned when a reference FASTA is available"

  - id: D4
    param: sampleSubset
    kind: optional
    source: user
    ask: "Should the output be restricted to particular samples?"
    default: "all samples"

  - id: D5
    param: regionSubset
    kind: optional
    source: user
    ask: "Should records be restricted to particular regions?"
    default: "whole file"

  - id: D6
    param: outputFormat
    kind: optional
    source: upstream
    ask: "Which container should the result be written in?"
    default: "bgzip-compressed VCF with an index"

  - id: D7
    param: threads
    kind: never_ask
    source: data
    reason: "Compression threads affect runtime only, not the records"
    default: "min(4, available_cores)"

# Typical post-calling workflow: normalize → filter → annotate → extract
bcftools norm -d any -f reference.fa variants.vcf.gz \
  | bcftools filter -i 'QUAL>20 && DP>10' \
  | bcftools annotate -a dbSNP.vcf.gz -c ID \
  | bcftools view -O z -o final.vcf.gz

# Index the output
bcftools index -t final.vcf.gz

# Count variants at each stage
bcftools stats final.vcf.gz | grep "^SN"

# VCF → compressed BCF (fastest format for piping)
bcftools view -O b -o variants.bcf variants.vcf

# BCF → VCF (for human-readable output)
bcftools view -O v -o variants.vcf variants.bcf

# VCF → bgzipped + indexed (standard archive format)
bcftools view -O z -W -o variants.vcf.gz variants.vcf
# -W automatically creates .tbi index after writing

# Extract specific samples
bcftools view -s sample1,sample2 -O z -o subset.vcf.gz variants.vcf.gz

# Exclude samples (prefix with ^)
bcftools view -s ^outlier_sample -O z -o cleaned.vcf.gz variants.vcf.gz

# Extract by region (fast; requires index)
bcftools view -r chr1:1000000-2000000 variants.vcf.gz -O v -o chr1_region.vcf

# Streaming pipeline: no intermediate files
samtools mpileup -Ou input.bam | bcftools call -m -Oz -o calls.vcf.gz

# Expression-based filter (include)
bcftools filter -i 'QUAL>20 && DP>10' variants.vcf.gz -O z -o filtered.vcf.gz

# Expression-based filter (exclude)
bcftools filter -e 'QUAL<10 || DP<5' variants.vcf.gz -O v -o filtered.vcf

# Soft filter: mark but keep (sets FILTER field to label)
bcftools filter -s LowQual -e 'QUAL<20' variants.vcf.gz -O z -o soft_filtered.vcf.gz
# Variants with QUAL<20 get FILTER="LowQual"; others get FILTER=PASS

# Keep only PASS variants
bcftools view -f PASS variants.vcf.gz -O z -o pass_only.vcf.gz

# SNP-only output
bcftools view --type snps variants.vcf.gz -O z -o snps.vcf.gz

# Indel-only output
bcftools view --type indels variants.vcf.gz -O z -o indels.vcf.gz

# Filter by allele frequency and depth
bcftools filter -i 'AF>0.1 && DP>20 && MQ>40' variants.vcf.gz -O z -o confident.vcf.gz

# Remove SNPs within 3 bp of indels
bcftools filter --SnpGap 3 variants.vcf.gz -O z -o gapfiltered.vcf.gz

# Extract chrom, position, ref, alt, quality
bcftools query -f '%CHROM\t%POS\t%REF\t%ALT\t%QUAL\n' variants.vcf.gz > variants.txt

# With header row (-H adds #-prefixed column names)
bcftools query -H -f '%CHROM\t%POS\t%REF\t%ALT\t%QUAL\n' variants.vcf.gz > variants.tsv

# Per-sample genotypes and allele depths
bcftools query -f '[%SAMPLE\t%GT\t%AD\n]' variants.vcf.gz > genotypes.txt
# Output: sample1  0/1  25,18  (ref_depth,alt_depth)

# Rare variants (AF < 1%)
bcftools query -i 'AF<0.01' -f '%CHROM\t%POS\t%REF\t%ALT\t%AF\n' \
    variants.vcf.gz > rare_variants.txt

# Count variants per chromosome
bcftools query -f '%CHROM\n' variants.vcf.gz | sort | uniq -c | sort -rn

# Extract genotype matrix across all samples
bcftools query -f '%CHROM:%POS\t[%GT\t]\n' -H variants.vcf.gz > genotype_matrix.tsv

# Merge: join VCFs from DIFFERENT sample sets (same variants)
bcftools merge sample1.vcf.gz sample2.vcf.gz sample3.vcf.gz \
    -O z -o cohort.vcf.gz

# Merge with auto-indexing and threading
bcftools merge -O b -W --threads 4 sample*.vcf.gz > cohort.bcf

# Concat: join VCFs from SAME sample set (different chromosomes or batches)
bcftools concat chr1.vcf.gz chr2.vcf.gz chr3.vcf.gz -O z -o full.vcf.gz

# Concat with overlap handling (from batched calling)
bcftools concat -a --threads 4 batch*.vcf.gz -O z -o concat.vcf.gz

# Pipeline: merge → filter → normalize
bcftools merge sample1.vcf.gz sample2.vcf.gz \
    | bcftools filter -i 'QUAL>20' \
    | bcftools norm -d any -f genome.fa \
    | bcftools view -O z -o merged_clean.vcf.gz
bcftools index -t merged_clean.vcf.gz

# Extract genotype matrix from merged cohort
bcftools merge cohort*.vcf.gz | bcftools query -f '[%GT\t]\n' > gt_matrix.tsv

# Add rsIDs from dbSNP
bcftools annotate -a dbSNP.vcf.gz -c ID variants.vcf.gz -O z -o rsid_annotated.vcf.gz

# Annotate with BED file (adds gene names)
bcftools annotate -a genes.bed.gz \
    -h <(echo '##INFO=<ID=GENE,Number=1,Type=String,Description="Gene name">') \
    -c CHROM,FROM,TO,GENE \
    variants.vcf.gz -O z -o gene_annotated.vcf.gz

# Remove unwanted INFO fields
bcftools annotate -x INFO/AC,INFO/AN,INFO/MQ variants.vcf.gz -O v -o stripped.vcf

# Normalize: left-align indels, split multi-allelic records
bcftools norm -f reference.fa variants.vcf.gz -O v -o normalized.vcf

# Split multi-allelic sites into separate records
bcftools norm -m -any variants.vcf.gz -O v -o split.vcf

# Deduplicate overlapping records
bcftools norm -d any variants.vcf.gz -O z -o deduped.vcf.gz

# Full normalize pipeline
bcftools norm -m -any variants.vcf.gz | \
    bcftools norm -d any -f reference.fa | \
    bcftools view -O z -o normalized_split.vcf.gz

# Full VCF statistics report
bcftools stats variants.vcf.gz > qc.stats.txt

# Extract Summary Numbers section only
grep "^SN" qc.stats.txt | cut -f3,4
# number of records:    45231
# number of SNPs:       38941
# number of indels:     6290
# ...

# Per-sample stats (PSC = per-sample counts)
bcftools stats -s - variants.vcf.gz | grep "^PSC" > per_sample.txt
# cols: id  sample  hom_RR  het  hom_AA  ts  tv  indel  missing  singleton

# Transition/transversion ratio (genome-wide QC)
bcftools stats variants.vcf.gz | grep "Ts/Tv"
# Ts/Tv ratio: 2.06 (healthy WGS; <1.8 or >2.2 suggests quality issues)

# Check for sample contamination (F-statistic per sample)
bcftools stats -s - variants.vcf.gz | grep "^PSC" | awk '{print $2, $9}'
# sample  F_missing (high = poor sample quality)

# Variant calling (mpileup → call pipeline)
samtools mpileup -Ou -f genome.fa *.bam | bcftools call -m -v -Oz -o calls.vcf.gz
bcftools stats calls.vcf.gz | grep "^SN"

-O v  → VCF text (uncompressed)       default for human inspection
-O z  → bgzipped VCF (.vcf.gz)        standard for archiving
-O b  → binary BCF (compressed)        fastest for piping
-O u  → binary BCF (uncompressed)      fastest output (no compression)

# INFO fields (one value per variant)
QUAL>20          # quality score
DP>10            # total depth
AF<0.05          # allele frequency
MQ>40            # mapping quality

# FORMAT fields (per-sample; use [] to iterate)
GT=="1/1"        # homozygous alternate
AD[1]>5          # alt allele depth > 5
GQ>20            # genotype quality

# Combined
QUAL>20 && DP>10 && AF>0.01
(GT=="0/0" || GT=="1/1") && GQ>30

#!/bin/bash
VCF="raw_calls.vcf.gz"
REF="reference.fa"
DBSNP="dbSNP_hg38.vcf.gz"
FINAL="variants_filtered_annotated.vcf.gz"

# 1. Normalize: left-align indels, split multi-allelic, deduplicate
bcftools norm -m -any $VCF \
    | bcftools norm -d any -f $REF \
    | bcftools view -O z -o normalized.vcf.gz
bcftools index -t normalized.vcf.gz

# 2. Filter by quality and depth
bcftools filter -i 'QUAL>20 && DP>10' normalized.vcf.gz \
    | bcftools filter --SnpGap 3 \
    | bcftools view -f PASS -O z -o filtered.vcf.gz
bcftools index -t filtered.vcf.gz

# 3. Annotate with rsIDs
bcftools annotate -a $DBSNP -c ID filtered.vcf.gz -O z -o $FINAL
bcftools index -t $FINAL

# Report variant counts
echo "Final variant count:"
bcftools stats $FINAL | grep "number of records"

#!/bin/bash
SAMPLES=(sample1 sample2 sample3 sample4 sample5)

# 1. Ensure all VCFs are indexed
for s in "${SAMPLES[@]}"; do
    bcftools index -t ${s}.vcf.gz
done

# 2. Merge into cohort VCF (only sites present in ALL samples: -m none)
bcftools merge -m none "${SAMPLES[@]/%/.vcf.gz}" \
    -O z --threads 8 -o cohort.vcf.gz
bcftools index -t cohort.vcf.gz

# 3. Filter: PASS, SNPs only, MAF > 1%
bcftools view -f PASS --type snps cohort.vcf.gz \
    | bcftools filter -i 'AF>0.01 && AF<0.99' \
    | bcftools view -O z -o cohort_snps_filtered.vcf.gz

# 4. Extract numeric genotype matrix (for plink/R/Python)
bcftools query -H \
    -f '%CHROM\t%POS\t%REF\t%ALT\t[%GT\t]\n' \
    cohort_snps_filtered.vcf.gz > genotype_matrix.tsv
echo "Genotype matrix: $(wc -l < genotype_matrix.tsv) variants x ${#SAMPLES[@]} samples"