Genetics and Molecular Biology · Skim version

Probability, Crosses and the Chi-Square Test

About 1 min to skim · 1 figure · Rewritten on

Read the full lesson → Skim the whole course →

The guiding idea

A 9:3:3:1 ratio predicts 312.75 round yellow seeds among Mendel's 556, and he counted 315. How far off can a count be before chance stops being a good explanation?

A cross predicts probabilities, and a probability predicts a spread of counts, not one exact count, so real offspring seldom match the ratio exactly.

From the film: A Ratio Predicts a Spread

  • A cross predicts probabilities, and real counts scatter around them.
  • For independent steps, multiply along a path: round and yellow is 3/4 × 3/4 = 9/16.
  • When several different outcomes count, add their paths.
  • These rules hold for unlinked genes; genes close together on one chromosome tend to be inherited together, and multiplying their single-gene chances fails.
  • Counts seldom match a ratio exactly; the question is how far a count can stray by chance.

From the film: How Far Is Too Far?

  • Chi-square scores how far counts lie from a ratio, judging each miss against its expected count.
  • If the ratio holds, chi-square averages its degrees of freedom: three for four classes.
  • With 3 degrees of freedom, only 5% of chance results score above 7.81.
  • The p-value is the share of chance results at least as far off as yours, computed as if the ratio and the other assumptions held.
  • A large p-value gives no reason to reject the ratio; it does not prove it.
  1. Two rules: and means multiply, or means add

    AI-generated illustration: three panels of chromosome schematics. Genes A and B on different chromosome pairs. Genes A and B far apart on one pair, with a crossing-over point between them. Genes A and B close together on one pair, bracketed as inherited together, with a crossover far from them.
    Genes on different chromosomes assort independently, and genes far apart on one chromosome behave almost as if unlinked, so the product rule applies. Genes close together on one chromosome tend to be inherited together, so their single-gene chances cannot simply be multiplied. AI-generated illustration
  2. Many genes at once: multiply along the branches

  3. Families: the binomial counts the birth orders

  4. Is the deviation chance? The chi-square test

  5. Going deeper

Read the full lesson → Skim the whole course →