Primer design
Our main goal was to cover the broadest possible spectrum of main phyla
present in the diet and of parasites of birds. We used a dataset of 5000
sequences selected randomly from SILVA SSU v132 reference and
non-redundant dataset (Quast et al., 2013), downloaded via the functionobisilva of Obitools software (Boyer et al., 2016). The selection
process was based on sequence annotation of the Silva database including
1000 sequences for each phylum Phragmoplastophyta, Apicomplexa,
Arthropoda, Nematoda and Platyhelminthes. These were aligned in Geneious
v11.1.4 (https://www.geneious.com; Kearse et al., 2012) and the
resulting alignment formed the basis for the design of the primers.
Primer design was performed using Geneious v11.1.4. (Biomatters Ltd, New
Zealand) with the following parameters: primer length 18–27
nucleotides, melting temperature (Tm) 57–63°C and GC content 30–80%.
Next, we assessed primer pairs for their potential to form dimers or
alternative amplicons. Two approaches were implemented: 1) 100
mini-barcodes (amplicon size < 200 bp) were designed to cover
the sequences of Nematoda and Platyhelminthes phyla (2000 sequences;
Supplemental information: Table S1); and 2) 20 mini-barcodes (amplicon
size 100-200 bp) were designed using the full dataset of 5000 sequences
(Supplemental information: Table S2). Mini-barcodes were then selected
according to the following criteria.
- Barcodes appearing in both approaches.
- Barcodes covering a variable region (V1 to V9) of the 18S rRNA gene.
- Barcodes for which primers were designed to cover the full alignment
area. Primers covering the edge of the alignment area were discarded
as in this region there was no information for many sequences.
- In the case of barcodes generated with similar primers, we selected
the longer mini-barcode.
After this filtering, three mini-barcodes were selected, two
(MiniB18S_41 and MiniB18S_43) for the V3 region and one (MiniB18S_81)
for the V7-V8 region (Table 1).