Skip to main content
. 2024 Jul 3;41(7):msae112. doi: 10.1093/molbev/msae112

Table 1.

Summary of current methods to quantify variation in recombination rate and distributions. See Peñalba and Wolf (2020) for a detailed review on different approaches. CO and GC refer to crossovers and gene-conversion events, respectively. References to methods and/or empirical examples are provided below the table. Sample size indicates the minimum number of individuals required for meaningful characterization of variation

Method Description Pros Cons Sample size
Cytogenetic estimation: Directly visualizes chromosomes in gametocytes using immunostaining of meiotic proteins to identify COs, often targeting foci of DNA mismatch repair protein MLH1. Can identify number and distribution of DSBs and the length of the meiotic axis/synaptonemal complex (e.g. targeting RAD51, SYCP3) Direct observation of DSB and CO rates. Physical positions are determined e.g. in μm Requires invasive sampling of gametocytes and often limited to males. Cannot give fine-scale positions relative to sequence features ≥1 to 10 s individuals
Pedigree-based estimation: Integrates pedigree and genetic marker information (e.g. SNPs) to identify marker pairs separated by recombination in gametes transmitted from parents to offspring. Can estimate recombination in (i) individuals, using information on recombination positions in gametes; and (ii) populations, creating linkage maps measured in centiMorgans (cM), where 1 cM is a 1% chance that two loci are separated by a CO event per meiosis Uses existing data from genotyped pedigrees. Quantifies individual variation. Can potentially identify GC-events in whole-genome data Requires large sample sizes to capture enough COs. Resolution of recombination positions limited by marker density ≥100 to 1,000 s individuals
Gamete sequencing: Sequences single- or pooled-gamete samples to identify recombination positions based on the deviation from parental or consensus allele frequencies within the same gamete and/or on the same sequencing read High precision of potential recombination positions within single individuals Often limited to male gametes due to ease of sampling. High sequencing costs ≥1 to 10 s individuals
Chromatin immunoprecipitation sequencing (ChIP-Seq): Isolates gametocytes and sequences genomic locations where specific proteins are bound to DNA using immunoprecipitation. DSB positions can be mapped by targeting proteins that initiate meiotic DSB formation (e.g. RAD51, DMC1) Identifies the specific sites of DSB formation within single samples Requires invasive sampling of gametocytes and often limited to males. Difficult to verify if DSBs are repaired by CO or GC ≥1 to 10 s individuals
Population-based estimation: Uses whole-genome sequence data to estimate the population-scaled recombination rate (ρ), based on patterns of linkage disequilibrium and the coalescent model Estimates fine-scale, sex-averaged recombination patterns over 100 to 1000 s of generations Affected by demography and selection, but new methods (e.g. neural networks) may overcome this. Cannot distinguish CO and GC-events ≥10 to 30 individuals, including outgroups
Phylogeny-based estimation: Two main approaches: (i) leverages incomplete lineage sorting (ILS) in phylogenies to infer recombination breakpoints in ancestral branch of two extant sister species; or (ii) uses the footprint of GC-biased gene-conversion in substitution patterns to quantify relative fine-scale recombination rates on terminal branches of phylogenetic trees Fine-scale estimates averaged over long periods of time (>millions of years) using few genomes (i) Requires substantial ILS between targeted species. (ii) Requires substantial GC-biased gene conversion in the tree Genomes from >4 or 3 closely related species, respectively.