Comparing Pool‐seq, Rapture, and GBS genotyping for inferring weak population structure: The American lobster (Homarus americanus) as a case study

Abstract

Unraveling genetic population structure is challenging in species potentially characterized by large population size and high dispersal rates, often resulting in weak genetic differentiation. Genotyping a large number of samples can improve the detection of subtle genetic structure, but this may substantially increase sequencing cost and downstream bioinformatics computational time. To overcome this challenge, alternative, cost‐effective sequencing approaches, namely Pool‐seq and Rapture, have been developed. We empirically measured the power of resolution and congruence of these two methods in documenting weak population structure in nonmodel species with high gene flow comparatively to a conventional genotyping‐by‐sequencing (GBS) approach. For this, we used the American lobster (Homarus americanus) as a case study. First, we found that GBS, Rapture, and Pool‐seq approaches gave similar allele frequency estimates (i.e., correlation coefficient over 0.90) and all three revealed the same weak pattern of population structure. Yet, Pool‐seq data showed FST estimates three to five times higher than GBS and Rapture, while the latter two methods returned similar FST estimates, indicating that individual‐based approaches provided more congruent results than Pool‐seq. We conclude that despite higher costs, GBS and Rapture are more convenient approaches to use in the case of species exhibiting very weak differentiation. While both GBS and Rapture approaches provided similar results with regard to estimates of population genetic parameters, GBS remains more cost‐effective in project involving a relatively small numbers of genotyped individuals (e.g., <1,000). Overall, this study illustrates the complexity of estimating genetic differentiation and other summary statistics in complex biological systems characterized by large population size and migration rates.

Exploring a Pool‐seq‐only approach for gaining population genomic insights in nonmodel species

“…Our observations are in line with this expectation with the smaller sample sizes from the previously genotyped SNPs showing smaller values (Figure , x ‐axes) of frequency of the most common allele than the allele frequency estimates from Pool‐seq (Figure , y ‐axes) when close to 0.5 but the opposite when the frequency of the most common allele is close to 1 (compare blue regression lines to orange expected lines in Figure ). Dorant et al () observe somewhat better correlations of allele frequencies from Pool‐seq versus conventional genotyping by sequencing than we do. However, they have larger sample sizes ( n = 48) in both pools and individual sequencing and use the same individuals in both.…”

“…For both pairs, we observe on average somewhat higher F ST values with the individual genotyped SNPs ( x ‐axes; Figure ) than from Pool‐seq F ST values ( y ‐axes; Figure ). Previous work has used allele frequencies directly to compare Pool‐seq approaches to conventional individual genotyping (Dorant et al, ; Hivert, Leblois, Petit, Gautier, & Vitalis, ; Zhu, Bergland, González, & Petrov, ), and when we compare frequency of the most common allele in each of the SNP loci from the previous individually based genotyping versus those from the pools, we find very strong correlation in all four populations (Figure a–d); coefficients of determination ( r 2 ) are over 0.7 in all four populations and linear regression coefficients are well over 0.8 and close to 0.9 for each of the naturally sympatric populations. We suggest that observed differences (Figure ) are largely due to the small sample sizes of the previous individually genotyped dataset of only n = 18 individuals for each of the introduced populations and n = 30 for each of the natural, sympatric pair.…”

Copy number variants outperform SNPs to reveal genotype-temperature association in a marine species

“…GBS, ddRADseq) with a sequence capture step using DNA probes designed from known genomic sequences. Here, we used the same 9,818 targeted loci previously used for the American lobster and all the details about the wet protocol are described in Dorant et al (2019). All Rapture libraries were sequenced on the Ion Torrent p1v3 chip at the Plateforme d'analyses génomiques of the Institute of Integrative and Systems Biology (IBIS, Université Laval, Québec, Canada http://www.ibis.ulaval.ca/en/home/).…”

“…This threshold was fixed to maximize the expected locus read depth among samples and remove poorly sequenced samples. BWA mem (Li 2013) was used to map the resulting individual-based sequences to the Rapture reference catalog previously developed by Dorant et al (2019) that consisted of 9,818 targeted sequences in total. SNP discovery was done with stacks v.1.48 (Catchen, Hohenlohe and Bassham, 2013).…”

Multi-model seascape genomics identifies distinct environmental drivers of selection among sympatric marine species

“…Additionally, using a reduced representation sequencing approach such as RAD-seq means that portions of the genome remain unknown, and many adaptive loci may not be captured [75]. However, for our research question, a pooled RAD-seq approach is bene cial as it allowed us to maximize the number of individuals per location to obtain accurate population allele frequency estimates [41,77], as well as maximize the number of sample sites, both of which are essential for GEAAs [62,78], without the full cost of sequencing every individual. As this study is a rst step in elucidating the putative adaptive potential of coastal invertebrate species in this unique marine realm, further studies using more complete genomic sequencing strategies are needed to characterize the full breadth of selection processes.…”