# Three-Point Crosses: Gene Order and Double Crossovers

> A middle marker makes double crossovers visible; comparing observed and expected doubles reveals crossover spacing. This lesson works a three-point test cross from eight offspring classes to gene order, adjacent-interval distances, coincidence and interference, then asks why a centimorgan is not a fixed length of DNA and where crossovers in humans and mice tend to fall.

Course: Genetics and Molecular Biology (Mapping and gene function). Level: core university (third year). About 17 minutes.
Lesson page (link this one when citing): https://stemhop.com/lessons/three-point-mapping/

Builds on: Linkage and Recombination Mapping · Chromosomes and Meiosis: The Physical Basis of Mendel's Laws · Probability, Crosses and the Chi-Square Test

## What you'll learn

- Order three linked genes from a test cross by matching each double-crossover class to its closest parental and finding the one gene that differs.
- Calculate each interval's map distance with the double crossovers counted in both intervals, and explain why a two-point count across the outer genes comes out smaller.
- Calculate the expected number of double crossovers, the coefficient of coincidence and the interference, and state what positive and negative interference mean.
- Compare genetic distance with physical distance, using hotspots, PRDM9, sex differences and the absence of crossing over in Drosophila melanogaster males.

The first film showed why a two-gene count, as in [linkage and mapping](https://stemhop.com/lessons/linkage-and-mapping/), misses two crossovers on one chromatid, and how a middle marker exposes them. A **crossover** exchanges segments between two non-sister chromatids in meiosis (see [chromosomes and meiosis](https://stemhop.com/lessons/chromosomes-and-meiosis/)).

A **locus** is a gene’s place on a chromosome; its versions are **alleles**. A **test cross** mates the individual being studied with one homozygous recessive at every locus followed, so each offspring’s phenotype reads one gamete from the other parent.

> **The one idea.** A middle marker makes double crossovers visible; comparing observed and expected doubles reveals crossover spacing.
>
> *Scope:* a three-point test cross with scorable classes and short adjacent intervals. Unequal survival can upset the usual size ranking, and even three markers miss an even number of crossovers within one interval.

## Read the eight classes in a fixed order

The first film sorted the eight classes into reciprocal pairs: **parentals** (no detected crossover), **single crossovers** in each interval, and **double crossovers** (one in each interval, usually smallest). Under independence, a double’s probability is the product of the two interval probabilities ([probability and crosses](https://stemhop.com/lessons/probability-and-crosses/)), and the product of two small fractions is smaller still.

**Worked problem.** A female heterozygous for recessive mutations p, q and r carries p q r on one chromosome and + + + (normal alleles) on the other. She is test-crossed to a p q r / p q r male. Assume every class survives equally well, phenotypes are scored without error, and her **phase**, the alleles together on each chromosome, is known. The 1,000 progeny are listed alphabetically, not in map order:

- p q r: 350
- 
  - 
    - +: 342
- 
  - q +: 53
- p + r: 55
- p q +: 96
- 
  - 
    - r: 92
- 
  - q r: 7
- p + +: 5

Total: 350 + 342 + 53 + 55 + 96 + 92 + 7 + 5 = 1,000.

**Step 1: find the parentals.** The two largest classes, p q r and + + +, match the female’s two chromosomes.

**Predict.** The provisional doubles are + q r and p + +. Match each to its closest parental. Which gene is in the middle?

> **Check your prediction.** - q r differs from p q r only at p, and p + + differs from + + + only at p. So p is in the middle, giving q–p–r. The alphabetical middle, q, is an outer gene.

*Interactive on the lesson page:* Three-point order finder: which gene is in the middle?

**Step 2: rewrite every class in map order,** q to p to r, and note where each switches from one parental pattern to the other.

- Parentals: q p r (350) and + + + (342)
- Singles between q and p: q + + (53) and + p r (55)
- Singles between p and r: q p + (96) and + + r (92)
- Doubles: q + r (7) and + p + (5)

Rarity is a clue, not proof: a class can be small because its offspring survive poorly, as Sturtevant reported for one wing mutation in 1913. With short intervals and every class surviving and scored equally well, take the reciprocal pair with the smallest combined count as the doubles. The map-order sort then labels the singles but cannot confirm that choice: any reciprocal pair taken as the doubles sorts just as neatly. If survival or scoring may differ, seek independent evidence.

**Step 3: calculate adjacent distances.** Map distance is measured in **[centimorgans](https://stemhop.com/methods/linkage-mapping/)** (cM), a frequency, not a length: for a short interval, 1 cM means about 1 gamete in 100 is recombinant. Count every double in both intervals:

> distance (cM) = 100 × (single crossovers in the interval + double crossovers) / total

- q to p: (53 + 55 + 7 + 5) / 1,000 = 120 / 1,000 = 12.0 cM
- p to r: (96 + 92 + 7 + 5) / 1,000 = 200 / 1,000 = 20.0 cM
- q to r on the map: 12.0 + 20.0 = 32.0 cM

**Predict.** If you now ignore p and score only q and r, what recombinant percentage will you obtain? Will it equal 32.0?

> **Check your prediction.** The doubles q + r and + p + carry parental q–r combinations, so only the 296 single crossovers count: 296 / 1,000 = 29.6%. The shortfall of 2.4 points is the 12 doubles counted once in each interval: 24 / 1,000.

Even 32.0 cM is a lower bound: two crossovers inside q–p or inside p–r, with no marker between them, still escape detection. That is why Sturtevant built maps from short adjacent intervals.

Twelve doubles turned up. Is that as many as chance predicts?

**Film transcript: Do Crossovers Act Independently?**

### Film transcript: Do Crossovers Act Independently?

The first film used a middle marker to expose double crossovers. Do those doubles occur as often as independent intervals predict? Keep the same blue and gold chromosomes, green middle marker, and red crossover points. Treat a crossover in each interval as an event. Under independence, multiply the two interval frequencies to predict double crossovers. The product fills an expected bar, marked schematic. Now compare an observed bar from the offspring with that expectation. Suppose the observed bar is shorter. Does that mean positive or negative interference?

Fewer doubles appeared than independence predicted. That is positive interference: a crossover in one interval is associated with fewer in its neighbour. The chromosome picture shows association, not a molecular cause. The coefficient of coincidence is observed doubles divided by expected doubles. Interference is the complement of coincidence.

Keep the pair together: coincidence measures what appeared; interference measures the departure from independence. Three comparisons organize the signs. Observed below expected gives positive interference. Equal bars give no interference. Observed above expected gives negative interference, an excess of doubles. That has been reported in some fungal and phage crosses. No single value holds everywhere. Interference usually weakens as the intervals lie farther apart. Observed doubles then approach the independent expectation. So the measurement belongs to one pair of intervals in one organism, not to a whole species. One cross gives an estimate, not a law.

A zero observed bar may reflect strong interference, a small sample, or missing classes. Check survival and scoring, then collect more offspring before claiming complete interference. Doubles have answered one question: how crossovers are spaced relative to each other. A second question is where along the DNA they fall. Answering it needs a second ruler. A genetic map records crossover frequency. A DNA ruler records physical length. Align them, and their spacing can disagree. Crossovers are not spread evenly, so equal DNA lengths can have different genetic distances. In humans and mice, many crossovers cluster in hotspots. A hotspot is a short DNA stretch with more recombination than nearby DNA.

The text below explains how hotspot positions are chosen and why that scope is not universal. Maps also change with the gamete maker and chromosome region. Female and male maps can differ. Some regions suppress crossovers. These patterns move the genetic marks without moving the underlying DNA sequence. Return to the two bars. Their comparison turns a hidden event into a spacing test, with scope limited to this interval pair. Comparing observed and expected doubles reveals crossover spacing.

From the film

- Comparing observed and expected doubles reveals crossover spacing.
- Under independence, expected doubles come from multiplying the adjacent interval frequencies.
- Coincidence is observed divided by expected; interference is one minus coincidence.
- Coincidence and interference describe one pair of intervals in one organism, not a species constant.
- Genetic distance reflects crossover frequency, not a fixed length of DNA.

## Compare observed doubles with chance

The second film named the **coefficient of coincidence**, observed doubles over the number expected under independence, and **interference**, one minus coincidence.

**Predict.** The interval fractions are 0.12 and 0.20. Among 1,000 offspring, how many doubles does independence predict, and is that above or below the 12 observed?

> **Check your prediction.** > expected doubles = 0.12 × 0.20 × 1,000 = 24
> >
> > coincidence = 12 / 24 = 0.50
> >
> > interference = 1 − 0.50 = 0.50
>
> Half the expected doubles appeared. This is **positive interference**: a crossover in one interval is associated with fewer crossovers in the neighbouring interval.

Interference of 1 means no doubles at all; a coincidence above 1 is negative interference, an excess of doubles, not an impossible result. These values describe one pair of intervals in one organism, not a species constant: interference tends to be strongest across short neighbouring intervals, and for intervals far apart, coincidence approaches 1. This is how visible doubles become a measure of crossover spacing.

### Design problem

**Predict.** Another three-point test cross has 500 progeny, intervals of 6 cM and 10 cM, and no double-crossover progeny. A classmate concludes that interference here is complete. Using the independent expectation as a Poisson mean, calculate P(0), then repeat for 5,000 progeny. Is the classmate justified?

> **One way to do it.** Independence predicts 0.06 × 0.10 × 500 = 3 doubles. A Poisson count with mean 3 is a good approximation for a rare event in many trials:
>
> > P(0) = e−3 = about 0.05
>
> The exact binomial, (1 − 0.006)500, gives 0.049. So even without interference, about one cross in twenty this size shows zero doubles. Zero is suggestive, not strong evidence, and cannot distinguish complete interference from interference of 0.5, which predicts 1.5 doubles.
>
> Score more progeny. With 5,000, independence predicts 30 doubles and P(0) = e−30, about 10−13. If none appear, interference is close to complete over these intervals; about 15 would indicate interference near 0.5. Also check that double-crossover classes survive and can be scored. Interference is plausible here, since it is strongest over short adjacent intervals like these.

## Genetic distance is not DNA length

The second film’s two rulers disagree on spacing, as Sturtevant warned they might in 1913. In humans the genome-wide average is close to 1 cM per million base pairs, only a rough average. In humans and mice, most crossovers fall in **hotspots**, short DNA stretches with much more recombination than the DNA around them.

In 2010, three groups linked hotspot position to *PRDM9*, whose protein binds particular DNA sequences through zinc fingers and marks nearby sites where recombination starts (see [DNA repair and recombination](https://stemhop.com/lessons/dna-repair-and-recombination/)).

**Predict.** In humans or mice, if the zinc-finger array of PRDM9 changes, should the same DNA sites remain hotspots?

> **Check your prediction.** Not necessarily: the bound sequences change, so hotspots move. People carrying different zinc-finger variants differed in which hotspots they used across the genome.

*Figure (AI-generated illustration).* In humans and mice, PRDM9's zinc fingers bind particular sequences and mark histone H3 nearby, and Spo11 breaks near such sites start recombination. Change the zinc fingers and the hotspots move. Image: a PRDM9 protein with a row of zinc-finger lobes sits on a short gold hotspot stretch of DNA, marks a nearby histone, and Spo11 makes a break beside it; in a second panel, different zinc fingers sit on a different stretch of DNA and the gold hotspot has moved.

Not every animal works this way. Dogs lack working PRDM9, birds appear to lack the gene, and their hotspots tend to lie near promoter-like features such as transcription start sites and CpG islands. Budding yeast shows a similar placement and far more recombination per kilobase than humans.

Maps also differ by sex and region. Across the genome, the human female map is about 1.6 to 1.7 times the male map. *Drosophila melanogaster* males normally have no meiotic crossing over, although males of some species, including *D. ananassae*, do. That is why *D. melanogaster* mapping crosses use a triple-heterozygous female, as in the final fly question. Crossovers are suppressed near centromeres in many organisms, including Drosophila and Arabidopsis; in *D. melanogaster* females, rates also varied strongly along chromosomes and between strains.

Maps and sequence agree on order, not always on spacing. Use a genetic map to predict how often genes travel together; use sequence to measure how much DNA separates them.

## Going deeper: when the simple comparison changes

Sturtevant showed in 1913 that doubles were scarcer than chance predicted. In 1915 he named the effect interference, crediting Hermann Muller with suggesting the name.

Positive interference is common, but not universal. Negative interference has been reported in some fungal and phage crosses. In Arabidopsis, one 2018 study measured interference of 0.64 for one interval pair; raising the dosage of the meiotic protein HEI10 lowered it to 0.34. In a 2016 sequencing study of Drosophila female meioses, noncrossover gene conversions, which exchange a short patch without exchanging the flanking DNA, did not show interference.

---

From STEMhop (https://stemhop.com), free lessons by Christopher Deppmann. Lesson text © the authors. Practice questions: https://stemhop.com/lessons/three-point-mapping/
