DNA metabarcoding has been increasingly used to detail distributions of hundreds of species. Most analyses focus on creating molecular operation taxonomic units (MOTUs) from complex mixtures of DNA sequences, but much less common is use of the sequence diversity within these MOTUs. Here we use the diversity of COI haplotypes within MOTUs from a California kelp forest to infer patterns of population abundance, dispersal and population history from 527 species of animals and algae from 106 samples of benthic habitats in Monterey Bay. Using haplotypes as a unit we show fine-grained differences of abundance across locations for 15 species, and marked aggregation from sample to sample for most of the common species of plants and animals. Previous analyses could not distinguish these patterns from artefacts of amplification or sequence bias. Our haplotype data also reveal strong population genetic differentiation over small spatial scales for 48 species of red algae, sponges, and Bryozoa. Last, phylogenetic analysis of mismatch frequencies among haplotypes show a wide variety of demographic histories from recent expansions to long, stable population sizes. These analyses show that abundant, small bodied marine species that are often overlooked in ecological surveys can have strikingly different patterns of ecological and genetic structure leading to population, ecological, and perhaps adaptive differences between habitats. MOTU diversity data from the same sequencing efforts that generate species-level analyses, and can greatly increase the scope and value of metabarcoding studies.

This dataset is a reanalysis of DNA metabarcoding data from Shum et al (2019, https://doi.org/10.1111/1755-0998.13067) to extract haplotypes using JAMP (https://github.com/VascoElbrecht/JAMP) and carry out demographic genetic analysis for hundreds of kelp forest species.

HaploCOBB_table.xlsx

This file contains the raw output of haplotypes from JAMP, the filtered haploCOBB dataset, Aggregation analysis (Figure S2) and mismatch analysis (Figure 9).

Rscript.zip

This zipped folder contain the R code (haploCOBB__Rscripts.txt) to recreate Figures in the main manuscript including the data files.

PopGen.zip

This zipped folder contains the HaploCOBB population genetic datasets of 527 MOTUs in fasta format including filter1 and filter2 datasets. .out files contain the DNAsp output of all datasets.

haploCOBB19_pipeline.zip

This zipped folder contains the bioinformatic workflow (haploCOBB_pipeline.txt) to generate haplotypes using OBITools and JAMP, including the generation of population genetic datasets from DNA metabarcoding data. The information containing the samples, barcodes and primers are contained in file ngsfilter.txt used to demultiplex the data.