Quick Start¶
Synopsis¶
Pipeline¶
panmap runs five stages in sequence. By default it runs through consensus. Use --stop to stop earlier.
Single-sample genotyping¶
Place reads onto the pangenome, align to the closest reference, call variants, and generate a consensus:
This produces sample.bam, sample.vcf, and sample.consensus.fa.
Metagenomic abundance estimation¶
Estimate which lineages are present in a mixed sample:
Output: sample.mgsr.abundance.out
Partial pipelines¶
A plain run executes the full pipeline (place -> align -> genotype -> consensus)
and writes .bam, .vcf, and .consensus.fa. Use --stop to end earlier:
# Stop at placement (writes .placement.tsv only)
panmap ref.panman reads.fq --stop place -o sample
# Place and align, skip genotyping and consensus
panmap ref.panman reads.fq --stop align -o sample
The --stop place/align/genotype/consensus stages are single-sample. With --meta, only --stop index applies (build the .midx and stop); the later stages don't run.
Managing indexes¶
panmap builds the index automatically on first run and reuses it after. The path is
derived from the panman (not -o): a placement index at <panman>.idx and an MGSR
index at <panman>.midx under --meta.
# Build the index only, next to the panman (no reads needed)
panmap ref.panman --stop index # -> ref.panman.idx (placement)
panmap ref.panman --meta --stop index # -> ref.panman.midx (metagenomic)
# Build to a custom path with --index-out (output), then load it with --index (input)
panmap ref.panman --stop index --index-out /data/ref.idx
panmap ref.panman reads.fq --index /data/ref.idx -o sample
# Force a rebuild
panmap ref.panman reads.fq --reindex -o sample
Next steps¶
- Single-Sample Mode -- full walkthrough with examples
- Metagenomic Mode -- wastewater and aeDNA workflows
- CLI Reference -- all options