Read me file for Waraniak JM, Marsh TC, and Scribner KT, ms. 18s rRNA metabarcoding diet analysis of the predatory fish community across seasonal changes in prey availability. This read me file describes the datasets and R code used in the statistical analyses outlined in the above manuscript. For any further queries, contact Justin Waraniak (jmwara@umich.edu). Waraniak_18S_dietseqs_cleaned.fasta Format: FASTA Contains the final output from the mothur processing of the raw fastq sequences. Sequences have been filtered by quality, sequences with <2 bp changes have been combined, chimeras and singletons have been removed, and sequences have been rarefied to 2000 sequences per sample. Sequences are ordered by their total abundance in all samples. Each new sequence is denoted by a “>” followed by the ID from the Illumina sequencing platform. Following the ID is the OTU number assigned by mother “Otu#####”. The total number of sequences corresponding the the OTU follows the OTU number, separated by a “|”. Following another divider “|” after the sequence count are the sample IDs which contained sequences corresponding to this OTU with each sample separated by a “-“. Following all of the samples an OTU sequence appeared in, on a new line is the representative nucleotide sequence for the OTU. Waraniak_18S_ESU_Biomass.txt Format: Tab-delimited text Contains the nightly drift net catch biomass data for larval fishes and macroinvertebrates used to make Figure 2 and calculate Chesson’s selectivity index. The top row contains the column labels for each of the 15 ESU taxa names for the prey ESUs that appeared in the sequences extracted from predator diet samples. The first column contains the dates for each day drift biomass data was collected for this study. The values underneath each taxa column label represent the total catch biomass (in grams) on the date in the first column. The final two rows represent the average nightly biomass for each ESU during the drift (23/5-7/6) and post-drift (3/7-5/7) respectively. Waraniak_18S_OTUseq_counts.txt Format: Tab-delimited text Contains the raw sequence counts for each of the OTUs identified as a potential prey item in each predator diet sample. The first row contains the ESU taxa that was identified by NCBI blast from the representative sequence of each OTU. The second row contains the main column labels. The first column "group" contains information on the plate ID and plate position for each predator diet sample that was submitted for Illumina sequencing. The second column "Sample ID" contains information on the ID given to a sample during field collection and was used to link the plate position ID to the field data collected for that predator sample. The third column "Species" contains the three-letter species code used to identify which species a predator diet sample was collected from. The remaining columns each represent the number of reads found in each one of the predator diet samples. The label for each of these columns correspond to the OTU number found in Waraniak_18S_dietseqs_cleaned.fasta. Waraniak_18S_Preyseq_Props.txt Format: Tab-delimited text Contains the proportion of sequences for each ESU in a predator diet sample. The proportion was calculated using the values from Waraniak_18S_OTUseq_counts.txt. OTUs with the same taxonomic label (top row of Waraniak_18S_OTUseq_counts.txt) or which had taxonomic labels that fell within a larger taxonomic group with an OTU label were combined into a single column, resulting in the 40 ESUs outlined in the manuscript. If a sample came from a predator species that was also represented by one of the prey ESUs, the read count for the ESU in that sample was changed to 0. The remaining reads were then divided by the total potential prey reads in each diet sample. The sum of all ESU proportions for a given sample should equal 1. This file is used as the input to run the analyses in the Waraniak_18S_Analyses.R "Sample ID" = the unique ID given to a predator diet sample during field collection. "Species" = the three-letter code denoting the species of the predator a diet sample was collected from. Species names corresponding to the three-letter fish codes can be found in appendix 4 of the following link: http://wcsu.csu.edu/cerc/researchreports/documents/LakewideAssessmentPlanForLakeMichiganFishCommunities1998.pdf "Transect" = the sampling transect a predator diet sample was collected from (see Figure 1). "Substrate" = the predominant substrate type of the sampling transect a predator diet sample was collected from. "Period" = denotes whether a predator diet sample was collected during the relatively high prey abundance drift period or the low prey abundance post-drift period. "Date" = denotes the calendar date a predator diet sample was collected on. The remaining columns contain the proportion of reads from the prey ESU in the column label that were present in each of the predator diet samples. Waraniak_18S_Analyses.R Format: R Code Contains R code for performing the principal coordinates, principal components, permanova, and bootstrapping tests reported in the manuscript. Contains functions for computing the Schoener index of dietary overlap and Chesson's selectivity index. Contains code for generating Figures 4, 5, and 6.