Helping decision making for reliable and cost-effective 2b-RAD sequencing and genotyping analyses in non-model species

Genome composition and GC content influence loci distribution in reduced representation genomic studies

BACKGROUND

Genomic architecture is a key evolutionary trait for living organisms. Due to multiple complex adaptive and neutral forces which impose evolutionary pressures on genomes, there is a huge variability of genomic features. However, their variability and the extent to which genomic content determines the distribution of recovered loci in reduced representation sequencing studies is largely unexplored.

RESULTS

Here, by using 80 genome assemblies, we observed that whereas plants primarily increase their genome size by expanding their intergenic regions, animals expand both intergenic and intronic regions, although the expansion patterns differ between deuterostomes and protostomes. Loci mapping in introns, exons, and intergenic categories obtained by in silico digestion using 2b-enzymes are positively correlated with the percentage of these regions in the corresponding genomes, suggesting that loci distribution mostly mirrors genomic architecture of the selected taxon. However, exonic regions showed a significant enrichment of loci in all groups regardless of the used enzyme. Moreover, when using selective adaptors to obtain a secondarily reduced loci dataset, the percentage and distribution of retained loci also varied. Adaptors with G/C terminals recovered a lower percentage of selected loci, with a further enrichment of exonic regions, while adaptors with A/T terminals retained a higher percentage of loci and slightly selected more intronic regions than expected.

CONCLUSIONS

Our results highlight how genome composition, genome GC content, RAD enzyme choice and use of base-selective adaptors influence reduced genome representation techniques. This is important to acknowledge in population and conservation genomic studies, as it determines the abundance and distribution of loci.

New colonisers drive the increase of the emerging loggerhead turtle nesting in Western Mediterranean

The loggerhead sea turtle (Caretta caretta) is sensitive to climate change and is responding by colonising the Western Mediterranean. To understand the rapid nesting increase in recent years in Spain, we sampled 45 hatchlings from 8 nests between 2016 and 2019. We sequenced a mtDNA D-loop region, genotyped 2291 SNPs using 2bRAD and collected data on clutch size, hatching success, and incubation duration. We confirmed that the colonisation has a Mediterranean and Atlantic mixed origin and we detected that these nests were laid by different females, except for two nests within the same season. Our results suggest that the recent increase in nesting is due to an increase in the number of colonising individuals rather than females born in the same area returning to breed. We hypothesize that this increase in the number of colonisers results from successful conservation efforts, feminisation of the populations of origin and earlier sexual maturation. However, the percentage of offspring females produced in Spain suggests that future returning individuals will aid to the settlement of the new population. These results allow defining the current status of this colonisation although future efforts are needed to detect remigrants to confirm the establishment of a resident population.

2b or not 2b? 2bRAD is an effective alternative to ddRAD for phylogenomics

Restriction-site-associated DNA sequencing (RADseq) has become an accessible way to obtain genome-wide data in the form of single-nucleotide polymorphisms (SNPs) for phylogenetic inference. Nonetheless, how differences in RADseq methods influence phylogenetic estimation is poorly understood because most comparisons have largely relied on conceptual predictions rather than empirical tests. We examine how differences in ddRAD and 2bRAD data influence phylogenetic estimation in two non-model frog groups. We compare the impact of method choice on phylogenetic information, missing data, and allelic dropout, considering different sequencing depths. Given that researchers must balance input (funding, time) with output (amount and quality of data), we also provide comparisons of laboratory effort, computational time, monetary costs, and the repeatability of library preparation and sequencing. Both 2bRAD and ddRAD methods estimated well-supported trees, even at low sequencing depths, and had comparable amounts of missing data, patterns of allelic dropout, and phylogenetic signal. Compared to ddRAD, 2bRAD produced more repeatable datasets, had simpler laboratory protocols, and had an overall faster bioinformatics assembly. However, many fewer parsimony-informative sites per SNP were obtained from 2bRAD data when using native pipelines, highlighting a need for further investigation into the effects of each pipeline on resulting datasets. Our study underscores the importance of comparing RADseq methods, such as expected results and theoretical performance using empirical datasets, before undertaking costly experiments.

A 2b-RAD parentage analysis pipeline for complex and mixed DNA samples

Next-generation sequencing technology has revolutionized genotyping in many fields of study, yet parentage analysis often still relies on microsatellite markers that are costly to generate and are currently available only for a limited number of species. 2b-RAD sequencing (2b-RAD) is a DNA sequencing technique developed for ecological population genomics that utilizes type IIB restriction enzymes to generate consistent, uniform fragments across samples. This technology is inexpensive, effective with low DNA inputs, and robust to DNA degradation. Here, we developed a probabilistic genotyping-by-sequencing genetic testing pipeline for parentage analysis by using 2b-RAD for inferring familial relationships from mixed DNA samples and populations. Our approach to partial paternity assignment utilizes a novel weighted outlier paternity index (WOPI) adapted for next-generation sequencing data and an identity-by-state (IBS) matrix-based clustering method for pedigree reconstruction. The combination of these two parentage assignment methods overcomes two major obstacles faced by other genetic testing methods: 1) It allows detection of parentage when closely related or inbred individuals are in the alleged parent population (e.g., in laboratory strains); and 2) it resolves mixed DNA samples. We successfully demonstrate this novel approach by correctly inferring paternity for samples pooled from multiple offspring (i.e., entire clutches) in a highly inbred population of an East African cichlid fish. The unique advantages of 2b-RAD in combination with our bioinformatics pipeline enable straightforward and cost-effective parentage analysis in any species regardless of genomic resources available.

Low impact of different SNP panels from two building-loci pipelines on RAD-Seq population genomic metrics: case study on five diverse aquatic species

Background

The irruption of Next-generation sequencing (NGS) and restriction site-associated DNA sequencing (RAD-seq) in the last decade has led to the identification of thousands of molecular markers and their genotyping for refined genomic screening. This approach has been especially useful for non-model organisms with limited genomic resources. Many building-loci pipelines have been developed to obtain robust single nucleotide polymorphism (SNPs) genotyping datasets using a de novo RAD-seq approach, i.e. without reference genomes. Here, the performances of two building-loci pipelines, STACKS 2 and Meyer’s 2b-RAD v2.1 pipeline, were compared using a diverse set of aquatic species representing different genomic and/or population structure scenarios. Two bivalve species (Manila clam and common edible cockle) and three fish species (brown trout, silver catfish and small-spotted catshark) were studied. Four SNP panels were evaluated in each species to test both different building-loci pipelines and criteria for SNP selection. Furthermore, for Manila clam and brown trout, a reference genome approach was used as control.

Results

Despite different outcomes were observed between pipelines and species with the diverse SNP calling and filtering steps tested, no remarkable differences were found on genetic diversity and differentiation within species with the SNP panels obtained with a de novo approach. The main differences were found in brown trout between the de novo and reference genome approaches. Genotyped vs missing data mismatches were the main genotyping difference detected between the two building-loci pipelines or between the de novo and reference genome comparisons.

Conclusions

Tested building-loci pipelines for selection of SNP panels seem to have low influence on population genetics inference across the diverse case-study scenarios here studied. However, preliminary trials with different bioinformatic pipelines are suggested to evaluate their influence on population parameters according with the specific goals of each study.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12864-021-07465-w.