Data - Making yogurt with the ant holobiont

<b>Abstract</b>Milk fermentation has a rich history in which food culture, the environment, and microbes intersect. However, the biocultural origins of fermentation practices and microbes have largely been replaced by industrial processes. Here, we consider a historical fermentation originating from Turkey and Bulgaria – ant yogurt. We revisit the traditional practices and modern gastronomic applications that use red wood ants (<i>Formica rufa</i> group) to initiate milk fermentation. Subsequently, we characterize the ants and experimental ant-derived yogurts. We uncover that the ant holobiont, which consists of the ants and their microbes, contributes key acids and enzymes to fermentation. Metabarcoding and culturing revealed that lactic and acetic acid bacteria, including species related to conventional sourdough, originate from the live ants and proliferate in the milk. The ants and bacteria consequently introduce formic, lactic, and acetic acid, advantageous for yogurt acidification and coagulation. Last, proteases with the potential to act on casein may alter yogurt texture and are produced by the ants and bacteria. The ant holobiont thus catalyses fermentation akin to the microbial consortia in other ferments. Our findings highlight the value of integrating traditional, gastronomic, and biological frameworks to uncover the origins and applications of microbes for fermented foods.Data files for 'Making yogurt with the ant holobiont uncovers bacteria, enzymes, and acids for food fermentation'.Data curator: Veronica M. Sinotte###########################<br><b>File 1: pH_metadata.txt</b>This file contains the pH measures of all of the ant yogurts and negative controls.These measures were taken after the samples were thawed and sieved to remove ant body parts.<br><b>File 2: SCRUB_metadata.txt</b>This file contains information on the ant and yogurt samples for the microbiome metabarcoding.It is used in the analysis pipeline, when potential contaminant are being removed. It is formatted according to the package SCruB.The row names are the Sample IDs.'is_control' TRUE indicates whether a sample was a negative control, meaning a blank sample was extracted, prepped, and sequenced. FALSE identifies biological samples.'sample_type' indicates if it was an ant, yogurt, or type of control'sample_well' is NA, as there was no applicable plate information given the experimental design. The samples were extracted in individual Eppendorfs and sequenced out of house, where plate information was not provided.<br><b>File 3: Decontam_sample_abundances.txt</b>This file contains information on the ant and yogurt samples for the microbiome metabarcoding.Listed is the proportion of sequencing remaining for each sample after decontaminating with the SCRuB pipeline.'SampleID' refers to the experimental sample ID.'scr_out$p' refers to the proportion of reads remaining for each sample. Here the negative controls are removed.<br><b>File 4: load_metadata.txt</b>This file contains information on the ant and yogurt samples bacteria loads and bacteria loads of select bacterial families.'Ng_Bact_DNA' This the total amount of nanograms of bacterial DNA per sample, based on qPCR of the 16S rRNA gene'Hypothet_Ng_Bac_Starter_or_Yog' Is the nanograms of bacterial DNA per ant starter. To calculate this, the number of ants in the extraction (4) and the number of ants added to the starter (53 on avg) were considered. For the yogurt, the amount in the extracted sample (9g) was multiplied by 3.3 to reach the total amount of yogurt (30 g) per fermentation.'Ng_Bact_DNA_mL' Is the nanograms of bacteria DNA in a mL of yogurt sample'Acetobacteraceae' The biomass of Acetobacteraceae per sample was estimated by multiplying the relative abundance (extracted from the metabarcoding data) by the total bacterial biomass in the sample.'Bacillaceae' The biomass of Bacillacear per sample was estimated by multiplying the relative abundance (extracted from the metabarcoding data) by the total bacterial biomass in the sample.'Lactobacillaceae' The biomass of Lactobacillaceae per sample was estimated by multiplying the relative abundance (extracted from the metabarcoding data) by the total bacterial biomass in the sample.<br><b>File 5: seqtab.nochim.csv</b>This file contains raw data to build a phyloseq object for the bacterial microbiome analyses.This file contains the negative control samples that were used for decontaminationThe rows represent each Sample ID, the columns each ASV and the corresponding number of reads for each sample.<br><b>File 6: decont_samp.csv</b>This file contains raw data to build a phyloseq object for the bacterial microbiome analyses.This is the corrected number of reads for each sample after removing contaminants predicted by SCRuB.The rows represent each Sample ID, the columns each ASV and the corresponding number of reads for each sample.<br><b>File 7: taxa.csv</b>This file contains raw data to build a phyloseq object for the bacterial microbiome analyses.Each row is an ASV, and each column represents levels of classification for the ASV.<br><b>File 8: decont_seq_metadata.csv</b>This file contains the metadata for making a phyloseq object with the decontaminated samples for the bacterial microbiome analysis.Note that here the negative controls for the extractions and sequencing were removed.<br><b>File 9: bacterial_isolates.xlsx</b>This file contains the information on all of the bacteria isolated and identified from the ant and yogurt samplesIt also contains the Accession Numbers of the consensus sequence for each sample from GenBank.The second spreadsheet contains the information that went into the Sankey diagrams.<br><b>File 10: HPLC_data.xlsx</b><b>Carbon Assay plate data.xlsx</b>This file contains data from Biolog PM1 and PM2 assay on Fructilactobacillus sanfrancisensis. The carbon substrates for each plate set-up are provided. The measures of a change in reaction color, indicative of catabolism of the specific substrates, was measured over 36 hours. <br>