Predicting RAD-seq Marker Numbers across the Eukaryotic Tree of Life

Area of Science:

• Genomics and Bioinformatics
• Molecular Evolution

Background:

• Restriction Site Associated DNA sequencing (RAD-seq) is vital for generating genome-wide genotypic data in eukaryotes.
• Accurate prediction of genetic marker numbers is essential for successful RAD-seq study design, but challenging for species lacking reference genomes.
• Current estimation methods rely on genome size and restriction recognition sequence probabilities, which are often unknown for nonmodel organisms.

Purpose of the Study:

• To systematically survey restriction enzyme recognition sequence frequencies across eukaryotes.
• To develop predictive models for estimating genetic marker numbers for RAD-seq studies.
• To provide a valuable resource for designing RAD-seq and related genomic studies, particularly for nonmodel species.

Main Methods:

• Performed in silico surveys of recognition sequences for commonly used type II restriction enzymes across the eukaryotic tree of life.
• Investigated the relationship between enzyme recognition sequence frequencies and phylogenetic relatedness.
• Developed predictive models using genomic compositions and cleavage frequency data to estimate recognition sequence probabilities.

Main Results:

• Recognition sequence frequencies for restriction enzymes exhibit significant variability among eukaryotic taxonomic groups, largely driven by phylogenetic relatedness.
• Genome sizes can be accurately predicted from restriction enzyme cleavage frequency data targeting neutral genomic elements.
• Models based on genomic compositions effectively calculate recognition sequence probabilities across taxa, applicable to existing reduced representation data.

Conclusions:

• Phylogenetic relatedness is a key determinant of restriction enzyme recognition site frequencies in eukaryotes.
• The developed analytical pipeline, PredRAD, and associated databases provide crucial tools for estimating marker numbers and guiding RAD-seq study design.
• These resources enhance the feasibility and success of genome-wide genetic studies for a wide range of eukaryotic species, especially those without reference genomes.