The Workflow of RAD Sequencing

Reduced-Representation Genome Sequencing (RRGS) refers to the targeted sequencing of specific portions of the genome. This cutting-edge technology employs restriction endonucleases to enzymatically sequence genomic DNA, enabling high-throughput sequencing of the enzymatically sequenced segments.

The library construction method classifies RRGS into three main categories: Reduced-Representation Libraries (RRL), Restriction-site Associated DNA (RAD), and Genotyping-by-Sequencing (GBS). Notably, RAD and GBS stand out as the most widely adopted methods, with variants like 2b-RAD, dd-RAD, and SLAF refining and advancing these techniques in various aspects.

In the realm of RRGS, RAD sequencing holds a pivotal role. The following elucidates the methods and steps involved in RAD sequencing.

Extraction and Quality Control of DNA Samples

For optimal results, it is advisable to utilize a DNA extraction kit for extracting DNA from both parents and populations, following the standard procedure. In cases where DNA has been extracted using the traditional SDS or CTAB method, an additional extraction step is recommended to eliminate proteins and RNA.

• Sample Type: DNA samples should undergo agarose electrophoresis to ensure minimal or no degradation. The samples should be free from contamination, with NanoDrop analysis confirming a DNA concentration within the OD260/OD280 ratio range of 1.8 to 2.0.
• Sample Requirement: Each sample preparation necessitates 1 μg of the sample. For multiple preparations, the total sample requirement is calculated as the number of preparations multiplied by 1 μg.
• Sample Concentration: The individual sample concentration should be equal to or greater than 25 ng/μl, with the recommended concentration falling in the range of 100 to 200 ng/μl.
• Recommended Sample Size: When dealing with offspring individuals, it is advisable to have a sample size of at least 100 individuals.

Ensuring adherence to these guidelines will contribute to the extraction of high-quality DNA samples, essential for subsequent analyses.

Library Construction and Sequencing

To streamline the genome sequencing process for species lacking reference genomes, the initial step involves the judicious selection of a suitable restriction endonuclease for digesting the target genome. This choice can be informed by existing literature or by referencing genomic information from closely related species with sequenced genomes. Simultaneously, the selected endonuclease undergoes enzymatic pre-tests on the genome, with the subsequent choice of the most suitable endonuclease for further experiments based on the pre-test results. The following outlines the steps for library construction and sequencing:

• Preparation and Drying of Samples, Barcoding, and Joins: Ensure proper preparation and drying of samples, including the incorporation of barcodes and joints.
• DNA Digestion: Employ a selected restriction endonuclease or a combination of endonucleases to digest genomic DNA.
• Adapter Addition to Digested DNA Fragments:

Attach adapters at both ends of digested DNA fragments, where single-enzyme RAD contains a barcode sequence at one end and no barcode at the other, and double-enzyme RAD features barcode sequences at both ends.

• Sample Pooling and Purification: Combine samples and purify the pooled DNA.
• Fragment Selection: For single RAD, randomly interrupt fragments using ultrasonic waves for selection; for double RAD, electrophoretically cut and recover target-sized fragments directly.
• PCR Amplification and Fragment Enrichment: Conduct PCR amplification and enrichment, with single-enzyme RAD utilizing universal joints and PCR enrichment, and double-enzyme RAD using the third step of joint primer PCR amplification and enrichment.
• PCR Product Purification and Library Sequencing after Quality Control: Purify PCR products and subject the library to sequencing after quality control measures.

It's worth noting that sequenced fragments obtained from single and double digestion RAD differ. Fragments from single digestion RAD are non-oriented, aligning only on the side with the cleavage site. In contrast, fragments from double digestion RAD are oriented, aligning reads on both sides.

Tag Clustering

(1) Data Quality Control

Following the acquisition of the initial sequenced sequences (Sequenced Reads), a meticulous process of quality control ensued. This involved the filtering of splice sequences, polyN, polyA, and other undesired sequences, resulting in a refined dataset termed 'cleandata.'

(2) Sample Classification and RAD-Tag Clustering with Barcodes

Post quality control, the samples underwent classification using barcodes. Duplicate reads arising from PCR amplification were then systematically eliminated. Subsequently, the ustacks module of the Stack software facilitated the assembly of clusters. The tag clustering, essential for organizing and grouping similar sequences, was then executed based on the sequence similarities within the dataset.

Variant Identification

Following the exclusion of RAD-tags exhibiting high read length depth (>500), biparental RAD-tags underwent a comprehensive analysis through BLAST comparison. This process led to the identification of SNPs and genuine InDels (≥2 bp) from the comparison results. Subsequently, these variants were scrutinized within segregating populations, with only SNPs demonstrating consistent polymorphisms in both parental and population datasets being retained.

In the case of single-enzyme cleaved RADs, sequences from the opposite side of each RAD-tag were meticulously spliced to generate contigs. These contig sequences were also employed for the identification of SNPs and InDels. Furthermore, if a contig's sequence contained a Simple Sequence Repeat (SSR), it served as a valuable resource for the development of PCR markers.

Genetic Map Construction

The segregation ratios of all markers within segregating populations underwent thorough examination using the chi-square test. Only markers exhibiting segregation ratios aligning with the inheritance pattern of a single locus, and with deletion data of less than 40%, were considered for inclusion in the genetic map construction process. For instance, in the F2 population adhering to the 1:2:1 ratio (pure parent 1: heterozygous: pure parent 2) and the RIL population with a 1:1 ratio (pure parent 1: pure parent 2), markers meeting these criteria were selected.

To address biased segregating markers, only those with a minimum allele frequency surpassing a critical value in the population were retained, drawing insights from literature pertaining to the target species. The software Joinmap proved instrumental in facilitating the subsequent genetic map construction.

Gene/QTL Localization

(1) Localization of Quality Trait Genes

While RAD sequencing is less commonly employed for the localization of quality trait genes, it is advisable to include phenotypic data from both the F2 population and the F2:3 family lines. This approach allows for the identification of heterozygous genotypes associated with the target trait within the F2 monoculture. The phenotypic data from F2 can be treated as one marker, with the genotypic data from the remaining markers entered into Joinmap for mapping. However, for optimal results in localizing quality trait genes, the Bulk Segregant Analysis (BSA) strategy is recommended.

(2) QTL Localization

RAD sequencing finds more widespread application in QTL localization, necessitating robust population phenotypic data and a population of recombinant inbred lines with a minimum of F6 generations. A comprehensive two-point data collection over at least two years for each family line in the population is essential. This involves randomized block trials with three replicates per point per year. Inaccurate QTL localization results may arise with fewer replicates. Ideally, the target trait should exhibit high heritability, and its phenotype should be minimally influenced by environmental conditions. The choice of appropriate QTL localization software should align with the characteristics of the target population and the number of markers available for the QTL localization process.