GitHub
08/31/2026, 7:55 AMreftreename column to testdata/phyloplace_input.csv, for nf-core/phyloplace#78 — summarising placements per reference tree rather than only per profile.
The existing rows already fell into two groups placing onto a shared reference tree, so the column only labels them; no data files are added or changed.
| row | reference tree | taxonomy | reftreename |
| ------------ | ---------------------------- | ------------------ | ----------- |
| PF14720wo | PF14720_seed.ft.LGCAT.newick | none | (empty) |
| PF14720wt | PF14720_seed.ft.LGCAT.newick | gappa_taxonomy.tsv | PF14720 |
| PF14720mafft | PF14720_seed.ft.LGCAT.newick | gappa_taxonomy.tsv | PF14720 |
| nuc_hmmer | cyanos_16s.newick | none | cyanos16s |
| nuc_mafft | cyanos_16s.newick | none | cyanos16s |
That covers a group with a taxonomy, a group without one, and a row that opts out. PF14720wo is left ungrouped deliberately: it declares no taxonomy where the other two rows on its tree do, and a joint classification can only use one taxonomy file. Each group also mixes the hmmer and mafft alignment methods on purpose, since placements made with different alignment methods still have to merge onto the same reference tree — verified that they do.
README.md documents the change under the phyloplace section.
Backwards compatible: the extra column is ignored by pipeline versions that do not know about it, so this can merge ahead of the phyloplace PR that uses it.
Generated by Claude
nf-core/test-datasetsGitHub
08/31/2026, 8:36 AMGitHub
08/31/2026, 8:50 AM