Breaking RAD: An evaluation of the utility of restriction site associated DNA sequencing for genome scans of adaptation

Abstract

Understanding how and why populations evolve is of fundamental importance to molecular ecology. Restriction site-associated DNA sequencing (RADseq), a popular reduced representation method, has ushered in a new era of genome-scale research for assessing population structure, hybridization, demographic history, phylogeography and migration. RADseq has also been widely used to conduct genome scans to detect loci involved in adaptive divergence among natural Correspondence: David B. Lowry, Fax: 517-353-1926; dlowry@msu.edu. Correction noteThe original advance online paper contained two errors associated with the calculation of the median density of RAD-seq tags in the survey of recent RAD-seq genome scan studies (Table S1). The first error was in the size of the assembled stickleback genome, which was reported as 0.53 Gbp, but should have been 0.46 Gbp. The second error was an accidental inversion of terms. These two mistakes contributed to an erroneous statement in the original abstract that the median density of RAD-tags across recent studies "was one marker per 3.96 megabases." The statement has been revised to read: "was 4.08 RAD-tag markers per megabase." Following these corrections, changes were made in the abstract and elsewhere in the paper to reflect a modified interpretation of the results, though we note the main arguments in the article are unaffected. Other minor modifications to the paper were made based upon suggestions by the editors of Molecular Ecology Resources. This version of the article, published as an "accepted article" on 12 November 2016 under DOI 10.1111/1755-0998.12635 replaces the original version of the article published on 12 September 2016 under DOI 10.1111/1755-0998.12596.The idea for the manuscript was conceived collectively by all authors during an NSF National Institute for Mathematical and Biological Synthesis (NIMBioS) working group. All authors contributed to the writing of the manuscript.Supporting Information Additional Supporting Information may be found in the online version of this article: Appendix S1 Supplementary R scripts for Breaking RAD. Table S1 Recent (January 2015 to April 2016) genome scan studies, which used RAD-seq for genotyping. Andrews et al. (2016). Generally, RADseq methods produce DNA libraries for high-throughput sequencing using restriction enzymes that cut at specific motifs throughout the genome. RADseq markers come in the form of RAD-tags, which are short-read sequences adjacent to restriction enzyme cut sites. Because many polymorphic markers are produced by RADseq, it has frequently been used successfully for population genetic analyses, including assessment of population structure, hybridization, demographic history, phylogeography and migration (Catchen et al. 2013;Cavender-Bares et al. 2015;Combosch & Vollmer 2015;Qi et al. 2015). Markers generated by RADseq have also been quite useful for constructing linkage maps and identifying quantitative trait loci (QTL;Pfender et al. 2011;Houston et al. 2012;Weber et al. 2013;Laporte et al. 2015...

Urban colonization through multiple genetic lenses: The city‐fox phenomenon revisited

“…These results should be interpreted with caution, due to the small sample size, minimal overlap of specific outlier sites, and ongoing debate on the use of RADseq data for detecting genomic outliers (Catchen et al, 2017;Lowry et al, 2016;McKinney et al, 2017).…”

Predictable genome-wide sorting of standing genetic variation during parallel adaptation to basic versus acidic environments in stickleback fish

“…conclusions about parallelism at the genomic level (Berner and Roesti 2017;Lowry et al 2017;Haenel et al 2018; for exceptions, see Martin et al 2013;Lamichhaney et al 2016;Reid et al 2016;Yeaman et al 2016;Elgvin et al 2017). Second, has the greater ecological difference of the acidic than basic lakes from the ancestral habitat caused asymmetry in the selection of adaptive genetic variation?…”

“…SNPs were ascertained in the global freshwater pool (i.e., all basic and acidic populations combined), requiring a total read coverage between 150 and 2800 (the latter effectively filtering sequences from repeated elements), a minor allele frequency (MAF) superior to 0.05 across the pool, and a distance to the nearest polymorphism of at least 12 base pairs (effectively avoiding microindel stutter). A total of 253,451 SNPs passed these filters, yielding an approximate average resolution of 1 SNP per 2 kilobases (kb)higher than in any previous reduced-representation sequencing study (Lowry et al 2017). At these SNPs, we performed nucleotide counts for each freshwater population at an average read depth of 63× per population pool.…”

Recent advances in conservation and population genomics data analysis

“…A recent study testing the impact of data processing on population genetic inferences using RAD‐seq data observed large differences between reference‐based and de novo approaches in population genetic summary statistics, particularly those based on the site frequency spectrum (Shafer et al., 2016). In addition, the recent debate over the effectiveness of RAD‐seq for discovering loci under selection (Catchen et al., 2017; Lowry et al., 2016; McKinney, Larson, Seeb, & Seeb, 2017) has highlighted the importance of testing the extent of linkage disequilibrium (LD) over the genome, whenever possible, in order to assess the power of genome scans to detect selected loci (e.g., Kardos, Taylor, Ellegren, Luikart, & Allendorf, 2016). …”

“…Focusing on the choice of sequencing method, a particular point of discussion at the ConGen 2017 workshop was the recent set of papers addressing the limitations of RADseq to illuminate the genetic basis of adaptation (Catchen et al., 2017; Lowry et al., 2016; McKinney et al., 2017). The primary criticism raised by Lowry et al.…”