Yeast communities from Nova Scotia vineyards before and after spontaneous fermentation
Description
Vineyards were sampled in the Annapolis Valley, Nova Scotia and assessed by targeting the fungal ITS regions using both Illumina MiSeq and PacBio Sequal II systems to compare yeast communities on grapes before and after spontaneous fermentation. Bulk samples were provided from eight vineyards over three years (2018-2020) at time of harvest, with 1-2 years of sampling per vineyard. PacBio sequencing was included in one year (2019) of this sampling. Additional samples were taken in a fourth year (2021) comprising three within-vineyard replicates from seven of the same vineyards and an additional ninth, and were assessed by Illumina MiSeq. Only pre-fermentation musts were assessed of 2021 samples. The SEED2 software was used to process sequence data and the resulting, unedited, BLASTn tables are uploaded here. Curated spreadsheets filtered for yeasts with species cross-checked and names updated are also included. Datasets will be added as they are assessed. Results thus far show that pre-fermentation communities have a high diversity of basidiomycete yeasts, and some unconventional yeasts were found post-fermentation, such as Saccharomyces uvarum.
Files
Steps to reproduce
One sample of L'Acadie blanc grapes was taken per year and split into 4 technical replicates that were pressed to produce must, then allowed to complete spontaneous fermentation. Samples were taken from each replicate before and after fermentation, DNA was extracted, and sequenced using the Illumina MiSeq system covering the ITS2 region with forward (ITS86: GTGAATCATCGAATCTTTGAAC, Turenne et al. 1999) and reverse (ITS4: TCCTCCGCTTATTGATATGC, White et al. 1990) primers, as well as using the PacBio Sequel system covering the full ITS region. The SEED2 software was used for data analysis. Sequences were trimmed to the length at which the medium quality value fell below 28, individual sequences with a sequence quality lower than 30 or a base pair quality value lower than 10 were removed, sequences were clustered into OTUs by complete-link clustering (USEARCH) and the most abundant sequence from each cluster was identified as the top NCBI BLASTn hit. The codes in the 'sample' column of the BLASTn tables for the 2018-2020 sampling are composed as follows: the first 1-3 characters identify the vineyard, the penultimate character represents the stage of collection (M=must, F=ferment) and the final character is the number of the replicate (1-4). For the 2021 replicates, the first 1-3 characters identify the vineyard and the final character is the replicate. To produce the dataset of only yeasts, the taxonomy of sequences belonging to the genus Saccharomyces was updated by phylogenetic binning using publicly available sequences. Species assignment of all sequences making up more than 1% of any sample were confirmed by manual NCBI BLAST searches., and species named were updated. Yeasts were filtered from the full dataset according to taxonomic groups obtained by consulting current yeast taxonomy literature. The numbers in these matrices are number of sequences. The codes in column A of the matrices are composed as follows, as is appropriate for the matrix: the vineyard site, the year, the stage of collection (M=must, F=ferment) and then the number of the replicate.