ITS analysis

ITS needs an extra step: removing the primers with cutadapt.

Primer removal

ITS amplicons vary a lot in length; when an amplicon is shorter than the reads, the reads run into the opposite primer. So, with marker_gene: ITS, PEMA runs cutadapt on the raw read pairs, before fastp trims and merges them:

• from read 1, the forward primer and the reverse complement of the reverse primer;
• from read 2, the reverse primer and the reverse complement of the forward primer;

as in the DADA2 ITS workflow.

Set the primers you used in the cutadapt section of the parameters file (5’→3’; IUPAC ambiguity codes are allowed), e.g.

marker_gene: ITS

cutadapt:
  forward_primer: GATGAAGAACGYAGYRAA
  reverse_primer: CTBTTVCCKCTTCACTCG

Any other cutadapt option can be added to the section, without the leading --, e.g. minimum-length: 50. The cutadapt report of each sample, with the number of reads in which each primer was found, is in cutadapt/ in the output folder.

ASVs or OTUs

Set clustering_algo to swarm (ASVs) or vsearch (OTUs); see 16/18S analysis.

Taxonomy assignment

crest4 has no built-in ITS database; PEMA provides one built from UNITE (2025): set database: unite2025 and mount_crest_dbs to a folder of yours (in the crest section), and PEMA downloads it from Zenodo the first time it is used. You can also use another crest4 database (put it in mount_crest_dbs and set database to its name), or build one from the UNITE sequences yourself; see Training the CREST classifier.