Bulk Export (dump)
afquery dump exports allele frequency data to CSV. It supports filtering by region and disaggregating output by sex, technology, or phenotype group.
Basic Usage
Export all variants to stdout:
Write to a file:
AC > 0 filter
By default dump exports only variants with AC > 0. Variants at covered positions with no carriers are omitted. Use --all-variants to include AC=0 rows, or afquery query --locus to verify coverage at a specific position.
Filter by Region
Export a single chromosome:
Export a specific region:
Positions are 1-based, inclusive on both ends.
Sample Filtering
Apply the same sample filters as query:
afquery dump --db ./db/ \
--phenotype E11.9 \
--sex female \
--tech wgs \
--output diabetic_female_wgs.csv
Disaggregate Output
All three disaggregation modes work on the same principle: add stratified columns alongside the totals. They can be combined in a single command. Each group adds the same columns as the base block, suffixed with the group label; n_eligible_<label> counts the samples of that group that can be genotyped at the position, and is 0 where none can.
Base columns (always present):
n_eligible is the number of samples that can be genotyped at the position:
they pass the sample filters, their technology covers the position, and they
carry at least one copy of the locus. It counts samples, not alleles, so on
chrX non-PAR a cohort of 2 males and 2 females has n_eligible=4 and AN=6.
It is the same value query reports.
The genotype columns mean the same as in query:
N_HET and N_HOM_ALT are diploid genotypes (every chrM carrier is in
N_HOM_ALT), and N_HEMI counts the males carrying the allele on chrX outside
the PARs or on chrY. With --by-sex, N_HEMI_female is always 0.
N_NO_COVERAGE is always emitted but is 0 unless a coverage-evidence
filter is active (see Coverage Evidence).
AF and every AF_<group> column are left empty when the matching AN is 0,
i.e. when no eligible sample of that group has a usable call at the position.
Add separate columns for male and female:
Output columns:
chrom,pos,ref,alt,AC,AN,AF,n_eligible,N_HET,N_HOM_ALT,N_HEMI,N_HOM_REF,N_FAIL,N_NO_COVERAGE,AC_male,AN_male,AF_male,n_eligible_male,N_HET_male,N_HOM_ALT_male,N_HEMI_male,N_HOM_REF_male,N_FAIL_male,N_NO_COVERAGE_male,AC_female,AN_female,AF_female,n_eligible_female,N_HET_female,N_HOM_ALT_female,N_HEMI_female,N_HOM_REF_female,N_FAIL_female,N_NO_COVERAGE_female
Add separate columns per sequencing technology:
Output columns include AC_wgs, AN_wgs, AF_wgs, n_eligible_wgs, N_HET_wgs, N_HOM_ALT_wgs, N_HEMI_wgs, N_HOM_REF_wgs, N_FAIL_wgs, N_NO_COVERAGE_wgs, AC_wes_v1, AN_wes_v1, etc. (one group of ten columns per registered technology).
Add separate columns for specific phenotype groups:
Output includes AC_E11.9, AN_E11.9, AF_E11.9, n_eligible_E11.9, N_HET_E11.9, N_HOM_ALT_E11.9, N_HEMI_E11.9, N_HOM_REF_E11.9, N_FAIL_E11.9, N_NO_COVERAGE_E11.9, AC_I10, etc.
Parallelism
Use multiple threads for faster export:
Full Option Reference
See CLI Reference → dump.
Next Steps
- Cohort Stratification — systematic cross-group AF comparison on specific loci
- Sample Filtering — filter syntax shared across query, annotate, and dump
- Performance Tuning — thread tuning for large export jobs