constructing a panel_meta.csv file correctly

[AOCuff]

hello,

I created a panel_meta.csv file in order to construct a cross-sample panel for calling indels. I did this using the format outlined in the scan2 wiki.

But the scan2 config ran using the makepanel analyses failed to read the donor demographic information held within the panel_meta.csv file. To double check that the BAM files were successfully being read, I ran the scan2 config using the gatk call_mutations analyses instead, and this ran successfully. So, I am inclined to think the panel_meta.csv file was not formatted incorrectly.

FYI, I outline the panel_meta.csv file as follows...

cat ../01_Input/panel_metadata/PD64547c_panel_meta.csv
donor,sample,amp
PD64547,PD64547c_lo0007,PTA
PD64547,PD64547c_lo0045,PTA
PD64547,PD64547c_lo0088,PTA
PD45517,PD45517k,MDA
PD64547,PD64547a_lo0001,bulk

...and here is the error report from the verbose output:

Checking reference genome..
Checking dbSNP VCF..
Checking BAMs..
adding bulk
adding SC
adding bam
Checking panel metadata..
ERROR: some samples listed in --makepanel-metadata have no associated BAM file (specified by --bam, --sc-bam, or --bulk-bam). Please ensure that every BAM in the metadata table is present. Note that samples IDs are automatically determined from BAM files by the SM: tag in the first read group (@RG record).

Could you possibly share an example of the panel_meta.csv file used in your analyses? I was unable to locate it in the current scan2 repo.
Where might I find the base script responsible for reading the panel_meta.csv file to run the makepanel analyses? I understand that SM tags from the BAM files might be involved, and these are correctly identified in my data. I would like to know the specific parameters that the scan2 algorithm anticipates in order to format the panel_meta.csv appropriately.

FYI, the commands are run as follows:

# scan2 -d panel init
scan2 -d "$OUT_DIR" init

sc_args=()
for bam in "${sc_bams[@]}"; do
sc_args+=(--sc-bam "$bam")
done

scan2 -d "$OUT_DIR" config \
    --verbose \
    --analysis makepanel \
    --regions-file "$REGIONS" \
    --ref "$REF" \
    --dbsnp "$DBSNP" \
    --phaser eagle \
    --eagle-refpanel "$EAGLE_REF" \
    --eagle-genmap "$EAGLE_GENMAP" \
    --makepanel-metadata "$PANEL_META/${sample}_panel_meta.csv" \
    --bulk-bam "$BULK" \
    "${sc_args[@]}" \
    --bam "$DONOR2"

scan2 -d "$OUT_DIR" validate

Many thanks for sharing your experience with this.