Outputs

Every file panmap writes is prefixed with the output prefix (-o, or the reads filename by default). Indexes are the exception: they are written next to the PanMAN so they can be shared across samples.

Single-sample mode

Running the full pipeline (panmap ref.panman reads.fq -o sample) produces:

File	Contents
sample.placement.tsv	Best-matching node for each placement metric
sample.ref.fa	Closest reference genome (with .ref.fa.fai index)
sample.bam	Reads aligned to the closest reference (with .bam.bai index)
sample.vcf	Variants called against the closest reference
sample.consensus.fa	Sample consensus sequence

Stopping early with --stop <stage> produces only the files up to that stage.

.placement.tsv

One row per placement metric, giving the metric's best-scoring node:

metric                score       nodes
log_raw               74.238025   England/CAMC-141DBEA/2021|OU190639.1|2021-03-19
log_cosine            0.787143    England/CAMC-141DBEA/2021|OU190639.1|2021-03-19
containment           0.196456    England/CAMC-141DBEA/2021|OU190639.1|2021-03-19
weighted_containment  1.065240    England/CAMC-141DBEA/2021|OU190639.1|2021-03-19
log_containment       0.454292    England/CAMC-141DBEA/2021|OU190639.1|2021-03-19

The metrics score how well the sample's seeds match each node's seed set:

Metric	Meaning
log_raw	Log of the raw number of shared seeds
log_cosine	Log cosine similarity of the seed-count vectors
containment	Fraction of the sample's seeds contained in the node
weighted_containment	Containment with each shared seed weighted by 1/(node genome count), up-weighting seeds that are rare across genomes
log_containment	Log-scaled containment (the default placement score)

The reference for alignment and genotyping is the node chosen by log_containment.

.vcf, .bam, .consensus.fa

Standard formats: the BAM holds reads aligned to sample.ref.fa, the VCF holds variants relative to that reference, and the consensus FASTA applies those variants to the reference. All are viewable in IGV, samtools, bcftools, etc.

Metagenomic mode

Abundance (--meta)

File	Contents
sample.mgsr.abundance.out	Estimated proportion per haplotype

Tab-separated node_id and proportion, sorted by proportion (proportions sum to ~1.0):

USA/MA-CDCBI-.../2021|OK176722.1|2021-09-07     0.50052
USA/SEARCH-6918/2021|PP046785.1|2021-01-21      0.20476
Germany/IMS-10036-.../2021|OU071802.1|2021-02-03 0.15086

Filter and assign (--meta --filter-and-assign)

File	Contents
sample.mgsr.assignedReads.fastq	The reads that were assigned
sample.mgsr.assignedReads.out	Per node: number of reads assigned and their indices into the FASTQ
sample.mgsr.assignedReadsLCANode.out	Per LCA node: assigned-read counts and indices (a read's LCA node is the LCA of all nodes it was assigned to)

Adding --breadth-ratio also writes sample.mgsr.breadths.out (per node: observed vs. expected seed breadth and depth).

Indexes

Indexes are reused on later runs.

File	Built by	Purpose
<panman>.idx	default / --stop index	Placement index (single-sample)
<panman>.midx	--meta	Metagenomic (MGSR) index

Use --index-out <path> to write to a custom location and --index <path> to load one. --reindex forces a rebuild. See the Quick Start for details.