Genetic linkage and recombination

Concept

Genetic linkage and recombination

T-015Home BU-103Threads information · evolution

Statement

Nearby genes tend to be inherited together.

Why it matters

independent-assortment establishes that genes on different chromosomes assort independently, but most genomes carry far more genes than chromosomes, so many gene pairs necessarily share a chromosome. Linkage and recombination describes exactly how those genes behave — travelling together more often than not, but occasionally separated by crossing over — and it is what converts observed inheritance ratios into an actual, ordered physical map of a chromosome, the earliest form of genetic mapping.

sex-linkage's characteristic inheritance pattern for genes on the sex chromosomes is a special case of this same physical-linkage logic, applied to genes riding along with the sex-determining locus itself.

Hypotheses

Genes located on the same chromosome are transmitted together more often than the 1:1:1:1 ratio independent-assortment predicts, because they are physically attached parts of the same DNA molecule during most of meiosis.Without physical co-location, there would be no mechanical reason for two loci to depart from the independent-assortment baseline at all. Crossing over between homologous chromosomes during meiosis I can separate linked alleles onto different chromatids, and crossover probability between two loci increases, roughly, with the physical distance separating them.Without this dependence on distance, recombination frequency could not be used to infer relative gene order or spacing at all. Recombination frequency underestimates true physical distance once loci are far enough apart that double crossovers become likely, since a double crossover between two loci restores the parental allele combination and is invisible to a simple recombination-frequency count — recombination frequency saturates and becomes an unreliable distance estimator well before loci are extremely far apart.

Proof

\text{Two loci on the same chromosome are transmitted together as parental (non-recombinant) combinations in the majority of gametes, unless a crossover occurs between them.}

This is the direct behavioural consequence of physical linkage (Hypothesis 1). A

RF = \frac{\text{recombinant offspring}}{\text{total offspring}}

Recombination frequency is measured directly from a test cross between a double heterozygote and a doubly homozygous recessive individual, where offspring phenotype directly reveals which gametes were parental versus recombinant. A

RF(\%) \approx \text{map distance (cM), for closely spaced loci}

For loci close together, \(RF\) rises roughly linearly with physical distance (Hypothesis 2); one map unit, one centiMorgan, is defined as the distance corresponding to a 1% recombination frequency. A

\text{The locus pair with the largest pairwise } RF \text{ among a triplet must be the two outermost loci.}

Three or more loci on the same chromosome can be ordered entirely from pairwise \(RF\) values this way; a three-point test cross gives gene order plus approximate spacing (Worked examples). B

RF \to 50\% \text{ as loci become sufficiently far apart on the same chromosome (or are on different chromosomes).}

At that point, crossover, or independent orientation, effectively randomises co-transmission, and linked loci become experimentally indistinguishable from unlinked ones by recombination frequency alone. A

Result

RF(\%) \approx \text{map distance (cM)}, \text{ for loci close enough together that double crossovers are rare}

Reading. Measured recombination frequency between two linked loci translates directly, at short distances, into a genetic map distance.

Scope. Breaks down, underestimating true distance, for widely spaced loci owing to invisible double crossovers, and saturates at a maximum of 50%, at which point it can no longer distinguish loosely linked from fully unlinked loci.

Corollaries & converses

independent-assortment is recovered as the limiting case of this framework when \(RF=50\%\), whether because two loci are genuinely on different chromosomes or simply very far apart on the same one — "unlinked" and "very loosely linked" are, from \(RF\) alone, indistinguishable.
sex-linkage's inheritance pattern for genes on a sex chromosome is a special case of the same physical-linkage logic applied specifically to genes co-located with the sex-determining locus.
Converse: given a set of pairwise \(RF\) values among several linked loci, the loci can be ordered into a single, consistent linear genetic map, and the internal consistency of that ordering (Step4) is itself evidence supporting the chromosome theory of inheritance's claim that genes are arranged linearly along a physical chromosome.

Fails without

Drop physical co-location on one chromosome (Hypothesis 1): without genes sharing a single DNA molecule, there would be no mechanical reason for them to be transmitted together more often than independent-assortment's 1:1:1:1 baseline, and \(RF\) for every gene pair would simply sit at 50% regardless of measured genomic distance, exactly as independent-assortment already predicts for genes on different chromosomes.
Drop distance-dependence of crossover probability (Hypothesis 2): if crossover frequency between two loci were unrelated to physical distance, \(RF\) could not be used to infer gene order or spacing (Step3–4), and genetic mapping — the entire practical application of this result — would not be possible.

Common errors

Assuming an \(RF\) of 50% always means two loci are on different chromosomes, rather than recognising it can equally mean they are very far apart on the same chromosome (Corollaries).
Treating recombination frequency as exactly linear and always accurate at any separation; it is only approximately linear for closely spaced loci and underestimates true distance once double crossovers become likely (Hypotheses, t3).
Forgetting recombination frequency is measured from a test cross specifically because it directly reveals gamete genotype through offspring phenotype; using a cross between two heterozygotes makes recombinant and parental classes far harder to distinguish.
Confusing "linked" (physically on the same chromosome) with "always co-inherited"; linkage only ever means co-inheritance is more frequent than the independent-assortment baseline, not absolute.

Discussion

Thomas Hunt Morgan and his students, working with Drosophila in the early twentieth century, first documented genetic linkage and used recombination frequency to construct the earliest genetic maps. Alfred Sturtevant, then an undergraduate in Morgan's laboratory, produced the first genetic map in 1913 by exactly this logic, converting a table of pairwise recombination frequencies into a single, ordered linear map.

Recombination frequency and true physical (base-pair) distance are correlated but not identical, since crossover rate per unit physical length is not perfectly uniform along a chromosome — some regions (recombination hotspots) undergo crossing over more often than their physical length alone would predict, and others, near centromeres, show suppressed recombination — so genetic maps in cM and physical maps in base pairs, while both correctly ordered, are not simply proportional to one another throughout.

Common misconception: that recombination is a rare, exceptional event. Crossing over occurs at least once on most chromosome arms during almost every meiosis; recombination frequency measures specifically the probability of a crossover occurring between two particular loci, not the overall frequency of crossing over on the chromosome.

Worked examples

\text{Three-point test cross among loci A, B, C: } RF(A,B)=9\%,\ RF(B,C)=6\%,\ RF(A,C)=15\%

Since \(RF(A,C)\) is the largest of the three pairwise values, and approximately equals \(RF(A,B)+RF(B,C)\), gene order must be A-B-C, with B lying between A and C (Step4). A

\text{Map: A}\ \overset{9\,\text{cM}}{-}\ \text{B}\ \overset{6\,\text{cM}}{-}\ \text{C}

The pairwise distances combine additively here, consistent with double crossovers being rare enough over this range to be neglected (Hypotheses, t3). A

\text{gene order A-B-C, spacing 9 cM and 6 cM}

Reading. Pairwise recombination frequencies alone, with no direct physical observation of the chromosome, are sufficient to reconstruct gene order and relative spacing.

Scope. The same method extends to any number of linked loci, at the cost of an increasingly complex, multi-point test cross.

Problems

A test cross between a doubly heterozygous parent (alleles linked) and a doubly homozygous recessive parent produces 82 parental-type and 18 recombinant-type offspring out of 100 total. Compute the recombination frequency and approximate map distance.
Solution
\(RF = 18/100 = 18\%\), corresponding to an approximate map distance of 18 cM (Step3).
Explain why a recombination frequency of exactly 50% between two loci does not, by itself, prove they are on different chromosomes.
Solution
Loci far enough apart on the same chromosome experience crossover frequently enough that their co-transmission is effectively randomised, also giving \(RF\approx50\%\) (Step5); \(RF\) alone cannot distinguish this case from genuinely unlinked loci on different chromosomes (Corollaries).
Given \(RF(X,Y)=4\%\), \(RF(Y,Z)=10\%\), \(RF(X,Z)=13\%\), determine the most likely gene order.
Solution
\(RF(X,Z)=13\%\) is the largest pairwise value, so X and Z are the outermost loci with Y between them; the small discrepancy from strict additivity (\(4+10=14\%\) versus the observed 13%) reflects a rare double crossover between X and Z that goes undetected as a recombinant (Hypotheses, t3).