Genetics without probability is like physics without mathematics — you can describe what happens, but you can't predict it. When Mendel observed his 3:1 ratios in pea plants, he was seeing the product law in action, without knowing it. In this post we'll build the two foundational probability rules from the ground up, apply them to coin tosses and chromosomes, and then tackle one of the most satisfying problems in introductory genetics: calculating the odds that a single sperm carries an all-maternal set of chromosomes.
Work through the practice questions and homework at the bottom — they are exam-style problems designed to test whether you really understand the reasoning, not just the answer.
What is probability, and why does it matter in genetics?
Probability is a number between 0 and 1 that describes how likely an event is to occur. An event with probability 0 cannot happen; one with probability 1 is certain. Everything in between is a prediction — and that is exactly what genetic ratios are.
When we say a dihybrid cross produces offspring in a 9:3:3:1 ratio, we are really saying that each individual offspring has a 9/16 probability of expressing both dominant traits, a 3/16 probability of expressing only the first dominant trait, and so on. These ratios are not guarantees for small families — they are predictions that become accurate over large sample sizes.
Genetic ratios like 3:1 or 9:3:3:1 are probability statements. They predict what happens across many offspring, not in any single cross. Deviations in small samples are expected — they are the result of chance, not error.
The two probability laws
Two rules govern almost every probability calculation you will encounter in genetics. Understand these and you can solve the vast majority of Mendelian problems.
The Product Law
When two or more independent events occur simultaneously, the probability of them both occurring is the product of their individual probabilities.
The Sum Law
When a single outcome can be achieved in more than one mutually exclusive way, the overall probability is the sum of the probabilities of each way.
These two rules are not abstract — they show up every time you flip a coin, roll a die, or watch a homologous pair of chromosomes segregate at meiosis I.
Worked example: the two-coin toss
Before we apply these laws to chromosomes, let's build intuition with a simpler system. Toss a penny (P) and a nickel (N) at the same time. Each coin is independent — the outcome of one tells us nothing about the other. Each coin has a 1/2 chance of landing heads (H) and a 1/2 chance of landing tails (T).
All four outcomes — Product Law
| Outcome | Penny | Nickel | Calculation | Probability |
|---|---|---|---|---|
| Both heads | Heads | Heads | ½ × ½ | 1/4 |
| Penny tails, nickel heads | Tails | Heads | ½ × ½ | 1/4 |
| Penny heads, nickel tails | Heads | Tails | ½ × ½ | 1/4 |
| Both tails | Tails | Tails | ½ × ½ | 1/4 |
One head and one tail — Sum Law
Now suppose we want to know the probability of getting exactly one head and one tail — we don't care which coin shows heads. There are two ways this outcome can occur: PH · NT and PT · NH. Each has probability 1/4. Because these are mutually exclusive ways of achieving the same outcome, we add them:
The key genetics application: all-maternal sperm (n = 10)
Here is the problem that ties probability theory directly to meiosis. It is a favourite in genetics exams and interviews — and once you see why the answer is what it is, it will stick.
In an organism with a haploid number of 10, what is the probability that a sperm will be formed that contains all 10 chromosomes whose centromeres were derived from maternal homologs?
The biological setup
During meiosis I, homologous chromosome pairs line up at the metaphase plate. For each pair, one chromosome came from the mother (maternal homolog) and one came from the father (paternal homolog). When the pair separates, each chromosome goes to one pole — and which pole each one goes to is completely random and independent of all other pairs. This is the basis of Mendel's Law of Independent Assortment.
The 10 homologous pairs — each segregating independently
Step-by-step reasoning
-
1For each of the 10 homologous pairs, the probability that the maternal homolog goes into a particular cell (and later into the sperm) is 1/2.
-
2The segregation of each pair is independent of all other pairs — the chromosome in pair 1 doesn't influence what happens in pair 2, 3, or 10.
-
3Because the events are independent, we apply the Product Law — we multiply the individual probabilities together.
-
4P(all 10 chromosomes are maternal) = (½)¹⁰ = 1/1024 ≈ 0.00098
Notice what this also tells us: there are 2¹⁰ = 1,024 possible combinations of maternal and paternal centromeres in the gametes of this organism. All-maternal is just one of those 1,024 equally likely outcomes.
Interactive calculator — explore any haploid number
Drag the slider to see how the probability changes as haploid number increases. Try n = 23 for humans.
Humans have a haploid number of 23. That means 2²³ = 8,388,608 possible chromosome combinations in a single gamete — and the probability of producing an all-maternal sperm is only 1 in 8,388,608. This extraordinary genetic diversity is generated entirely by independent assortment, before even considering crossing over.
Practice questions
Test your understanding with these exam-style questions. Click an answer to see whether you're right — and read the explanation even if you get it correct.
Click to answer
In a monohybrid cross Aa × Aa, what is the probability that a randomly chosen offspring is homozygous recessive (aa)?
In a dihybrid cross AaBb × AaBb, what is the probability of an offspring that expresses exactly one dominant trait?
A fair coin is tossed three times. What is the probability of getting exactly two heads?
An organism has a haploid number of 3. How many different chromosome combinations are possible in its gametes?
Two genes, A and B, assort independently. For a cross AaBb × aabb, what is the probability that an offspring is AaBb?
Probability predictions are accurate for large sample sizes. A predicted ratio of 3:1 means that in a population of thousands of offspring, approximately three-quarters will show the dominant phenotype — not that exactly 3 out of every 4 will. In small families, chance deviations are entirely normal. We use the chi-square (χ²) test to determine whether an observed deviation is within the range expected by chance.
Take-home Problems
Complete these before your next session. Show full working for each calculation.
1. Using the product law, calculate the probability that in a human (haploid number = 23), a single sperm will contain all 23 chromosomes whose centromeres are maternally derived. Express your answer as both a fraction and a decimal. What does this tell you about genetic diversity?
2. In a trihybrid cross AaBbCc × AaBbCc, what fraction of offspring are expected to be (a) homozygous dominant for all three loci, and (b) heterozygous at all three loci?
3. P(event A) = ½ and P(event B) = ¼; the events are independent. Calculate: (a) P(A and B), (b) P(A or B, but not both). Show your working using the appropriate law for each part.
4. An organism has a haploid number of 5. List all possible gamete types based on whether each chromosome is of maternal (M) or paternal (P) origin. How many total combinations are there? What is the probability of each?
5. Reflection. Why do we emphasise that genetic ratios are most meaningful over large sample sizes? Give a real example from genetics history to support your answer. What statistical test would you use to evaluate whether observed ratios deviate from expected ratios?