Selective Breeding Virtual Lab: Artificial Selection

This selective breeding virtual lab follows the AP Biology artificial selection investigation with Wisconsin Fast Plants, and it covers the GCSE topic of selective breeding. Students count leaf hairs (trichomes) on 300 plants, keep the hairiest 10% as parents, breed them and compare the offspring with the parents' generation. They record the selection differential S and the response R, calculate realized heritability R/S, and run five generations in minutes instead of 200 days. A second run with very few parents shows why breeders worry about inbreeding.

Curriculum links

• AP Biology Topic 7.3 (artificial selection, EVO-1) and the AP Biology artificial selection investigation with Fast Plants trichomes.
• AQA GCSE Biology 4.6.2.3 (selective breeding): choosing parents with a desired characteristic over many generations, and the risk of inbreeding.
• NGSS MS-LS4-5 (technologies that change the inheritance of traits) and IB Biology D4.1 (natural selection).

Simulic is not affiliated with or endorsed by the College Board, AQA, the IB or any exam board.

Before the lab (5 min)

Ask students to commit to a prediction, on paper or as question 1 of the class link:

"You breed only from the hairiest 10% of the plants. Their mean is about 7 trichomes above the whole population's mean. What will the mean of their offspring be?"

Many students expect the offspring to match their parents exactly. Others think nothing will change in one generation.

Method in the simulation

The simulation opens on Darwin's finches, so students switch mode first.

1. Under the simulation, set Mode to Selective breeding. Keep Trait Fast Plants: leaf hairs, h² 0.70, Parents kept 10% and Select: highest.
2. Read the summary line below the controls: chosen 30/300, the population mean μ, the parents' mean μₛ, S = μₛ − μ and the predicted R = h²S.
3. Press Breed selected. The breeding record adds a row with S, h²S, the actual response R and R/S, the realized heritability.
4. Press Breed selected four more times. Record the mean, SD, F and V_A for each generation.
5. Press Reset, set h² to 0.30 and breed one generation. Compare R with the first run.
6. Set h² back to 0.70, set Parents kept to 1% and breed five generations. Compare F, V_A and the warnings with the 10% run.

Generation	Parents	Mean μ	SD	S	h²S	R	R/S	F	V_A (% of start)
0 → 1	30/300
1 → 2	30/300
…
4 → 5	30/300

Expected results

The simulation is random, so every class gets slightly different numbers. These ranges come from thousands of model runs, confirmed in the browser.

• Generation 0: mean about 8 trichomes per plant (7.5 to 8.6), SD about 4.
• First selection, top 10%, h² = 0.70: S is about 7.0 (6.1 to 7.9). The predicted R is about 4.9 and the actual R about 4.8 (3.7 to 6.0), so the offspring average about 12.8 trichomes. Realized heritability R/S is about 0.69 (0.56 to 0.83).
• Five generations (about 200 days): the mean reaches about 27 trichomes (25 to 30). F = 0.081 and V_A falls to about 60% of the start, so each step gets slightly smaller.
• h² = 0.30: the same S gives R of only about 2 (1.0 to 3.2), and R/S about 0.30.
• 1% kept (3 parents): F = 0.167 after one generation and 0.598 after five, V_A drops to about 25 to 30% of the start, the SD shrinks to about 2.8, and the sim shows an inbreeding warning.

Questions for students

1. (Prediction, asked again after the lab) Breeding from the hairiest 10%: what will the offspring mean be?
2. When you compare h² = 0.70 with h² = 0.30, what must stay the same?
3. First generation, top 10%, h² = 0.70: what is S?
4. From the same row, calculate the realized heritability R/S.
5. Why is keeping only the top 1% risky, even though it gives a bigger S?

Answers for teachers: (1) Higher than the population mean, but not as high as the parents' mean. (2) The share of parents kept and the population size. (3) About 7 trichomes (accept 5.8 to 8.3). (4) About 0.7 (accept 0.5 to 0.9). (5) With 3 parents the plants become closely related: F rises to about 0.6 in five generations and V_A falls to about a quarter, so later progress slows and inbreeding depression becomes likely.

Common misconceptions

• "Offspring inherit their parents' exact trait." Only the additive genetic part is passed on. The environmental part is not, so R is smaller than S.
• "Selective breeding creates new variation." It uses variation that already exists, and it uses some up: V_A and the SD fall each generation.
• "Plants change because they need more hairs." Nothing in the plants responds to need. The breeder decides which plants reproduce.

Extension

• Read the Selective breeding vs genetic modification card under the simulation and compare the time each method needs.
• Switch back to Natural selection and try Stabilising and Disruptive selection: the mean hardly moves, but the SD shrinks or grows.

FAQ

Does this replace growing Fast Plants?

No. The real investigation takes weeks and teaches plant care, counting and pollination. The simulation lets students plan it, practice the calculations and see five generations in one lesson. See virtual labs vs physical labs.

Why don't my numbers match my partner's?

Each population and each breeding step is random, like real plants. Compare the class mean of R/S with the h² you set.

Can I start the class in breeding mode?

Yes. On your class link, set the starting value Mode = Selective breeding, and choose the trait and the share of parents kept.

Related simulations and guides

Natural selection – changing allele frequency (peppered moth) Natural selection – rabbit population, mutations, wolves and food

For allele frequencies in a population, see the Hardy–Weinberg virtual lab. For more ideas, see interactive biology lesson ideas.