Conditional Probabilities — Compute conditional...

SECTION 1

Historical Context & Motivation

When Gregor Mendel crossed pea plants in the 1860s, he noticed predictable patterns in how traits like flower color and seed shape were passed from parents to offspring. He counted thousands of plants and realized that certain ratios — like 3 to 1 — kept appearing. To explain those ratios, scientists eventually turned to the mathematics of probability (the study of how likely events are to happen).

But here's the twist: in real life, we often already know something about an organism before we ask a probability question. For example, you might know a child looks healthy but wonder whether they carry a hidden disease allele. That extra piece of information changes the odds. The mathematical tool for updating probabilities when you have extra information is called conditional probability.

1763

Bayes' Theorem Published

Reverend Thomas Bayes showed how to update the probability of an event when new evidence is available. His theorem is the backbone of conditional probability.

1866

Mendel's Laws of Inheritance

Gregor Mendel published his experiments on pea plants, revealing dominant and recessive traits and predictable inheritance ratios.

1908

Hardy-Weinberg Principle

Godfrey Hardy and Wilhelm Weinberg described how allele frequencies stay constant in a population, connecting probability to population genetics.

1950s–1970s

Genetic Counseling Emerges

Doctors began using conditional probability to advise families about the chance their children could inherit genetic disorders like cystic fibrosis or sickle cell disease.

Today

DNA Testing & Bayesian Analysis

Modern genetic tests use conditional probability to interpret results, from newborn screenings to ancestry kits.

The central question this lesson addresses is: When you already know something about an organism — such as its appearance — how do you calculate the exact probability of its hidden genotype?

SECTION 2

Core Principles & Definitions

Before we dive into calculations, let's lock in the key ideas. You already know that an organism's genotype (its combination of alleles, like Bb) determines its phenotype (its observable trait, like brown eyes). In many cases, more than one genotype can produce the same phenotype. That's exactly where conditional probability steps in.

Simple Probability

The chance of an event without any extra information. Example: the probability of rolling a 4 on a die is 1/6.

Conditional Probability

The chance of an event given that something else is already known. Written as P(A | B), read 'the probability of A given B.'

Sample Space Reduction

When you learn new information, you eliminate impossible outcomes. Your sample space (the set of all possibilities) shrinks.

Carrier

An organism that is heterozygous (has one dominant and one recessive allele, like Aa) and shows the dominant phenotype but can pass the recessive allele to offspring.

Bayes' Theorem

A formula that flips a conditional probability around. It lets you calculate P(Genotype | Phenotype) when you know P(Phenotype | Genotype).

✦ KEY TAKEAWAY

Think of conditional probability like filtering search results. If you search "shoes" you get thousands of results. But if you add the filter "color = red," the list shrinks. Conditional probability works the same way: knowing an organism's phenotype is like applying a filter that removes impossible genotypes, and you recalculate the odds using only the possibilities that remain.

SECTION 3

Visual Explanation — Filtering the Punnett Square

The diagram below shows a standard monohybrid cross between two heterozygous parents (Bb × Bb). Normally we say the offspring have a 1/4 chance of being BB, a 2/4 chance of being Bb, and a 1/4 chance of being bb. But what if we already know the offspring shows the dominant phenotype? Then the bb square is impossible, and we only look at the remaining three squares.

The green-bordered squares show genotypes that produce the dominant phenotype. The red-dashed square (bb) is eliminated because it does not match the known phenotype. Of the 3 remaining squares, 1 is BB and 2 are Bb, so the conditional probability of being a carrier (Bb) is 2/3.

Notice the key move: we didn't change the genetics. We just changed which outcomes we consider. The condition — "this organism shows the dominant phenotype" — acts like a filter. It removes the bb possibility, and we recalculate using only the remaining three equally likely outcomes.

SECTION 4

Mathematical Framework

The formal definition of conditional probability looks like this. Don't worry — we'll break it down piece by piece with genetics examples.

CONDITIONAL PROBABILITY FORMULA

P(A | B) = P(A and B) / P(B)

P(A | B) = probability of event A, given that event B has occurred. P(A and B) = probability that both A and B happen together. P(B) = probability of event B happening at all.

In genetics, A is usually a genotype and B is usually a phenotype. Let's translate: "What is the probability that an offspring is Bb (event A), given that it shows the dominant trait (event B)?" Using the Bb × Bb cross from our diagram, P(Bb and dominant) = 2/4, because two of the four squares are Bb and both show the dominant trait. P(dominant) = 3/4, because three of the four squares show the dominant phenotype. Plugging in: P(Bb | dominant) = (2/4) ÷ (3/4) = 2/3.

GENETICS EXAMPLE

P(Bb | dominant) = P(Bb and dominant) / P(dominant) = (2/4) / (3/4) = 2/3

Among offspring that display the dominant phenotype from a Bb × Bb cross, two out of three are carriers.

MULTIPLICATION RULE FOR INDEPENDENT EVENTS

P(A and B) = P(A) × P(B) [if A and B are independent]

This rule is handy when you combine conditional probability with multi-step crosses. For instance, the probability that two independent carrier parents both pass the recessive allele is (1/2) × (1/2) = 1/4.

💡 Why Divide by P(B)?

Dividing by P(B) rescales everything so that the probabilities of the remaining outcomes add up to 1 (100%). It's like zooming in on just the slice of the pie chart that matters.

SECTION 5

Applying Conditional Probability to Pedigrees

A pedigree is a family tree diagram that tracks a trait through generations. When you see an unaffected individual in a pedigree and both parents are carriers, you immediately face a conditional probability question: given that this person is unaffected, what is the probability they are a carrier?

This pedigree shows two carrier parents (Aa × Aa) and their four possible offspring genotypes. The filled purple circle represents the affected individual (aa). The conditional probability box at the bottom shows the key result: among unaffected offspring, 2/3 are carriers.

This 2/3 result is one of the most important numbers in genetics problem-solving. Genetic counselors use it constantly. Whenever two carriers have a child who looks healthy, the chance that the healthy child is secretly a carrier is not 1/2 — it's 2/3. Many students get this wrong on tests because they forget to apply the condition!

Unconditional vs. conditional probabilities for offspring of two Aa parents

Genotype	Unconditional P	Shows Dominant?	Conditional P (given dominant)
AA (homozygous dominant)	1/4 = 25%	Yes ✓	1/3 ≈ 33.3%
Aa (carrier)	2/4 = 50%	Yes ✓	2/3 ≈ 66.7%
aa (homozygous recessive)	1/4 = 25%	No ✗ (eliminated)	0% (impossible)

SECTION 6

Worked Example — Cystic Fibrosis Carrier Risk

Cystic fibrosis (CF) is an autosomal recessive disorder. Both of Maria's parents are known carriers (Cc). Maria is healthy (she does not have CF). She wants to know: what is the probability she is a carrier? And if she has a child with a partner who is also a carrier, what is the probability their child will have CF?

Finding Maria's Carrier Probability and Her Child's Risk

Step 1 — List all possible genotypes from the crossMaria's parents are both Cc. The Punnett square gives four equally likely outcomes: CC (1/4), Cc (2/4), and cc (1/4).

Possible genotypes: CC, Cc, Cc, cc

Step 2 — Apply the condition (Maria is healthy)Since Maria does not have CF, she cannot be cc. We eliminate that outcome. Three equally likely genotypes remain: CC, Cc, Cc.

P(Maria is a carrier | healthy) = 2/3

Step 3 — Set up the partner's informationMaria's partner is a confirmed carrier (Cc), so P(partner is Cc) = 1.

Partner genotype: Cc (certain)

Step 4 — Calculate the child's risk using total probabilityThere are two scenarios. Scenario A: Maria is CC (probability 1/3). In a CC × Cc cross, 0% of offspring are cc. Scenario B: Maria is Cc (probability 2/3). In a Cc × Cc cross, 1/4 of offspring are cc.

Total P(child has CF) = (1/3 × 0) + (2/3 × 1/4) = 0 + 2/12 = 1/6 ≈ 16.7%

Step 5 — Interpret the resultThere is approximately a 1 in 6 chance that Maria and her carrier partner will have a child with cystic fibrosis. A genetic counselor would share this information so the couple can make informed decisions.

Answer: P(child has CF) = 1/6 ≈ 16.7%

SECTION 7

Strengths, Limitations & Common Mistakes

When conditional probability in genetics works well and when it doesn't

Strengths	Limitations
Gives precise, updated probabilities when you have partial information about phenotype.	Assumes complete dominance. Incomplete dominance or codominance requires modified analysis.
Essential for genetic counseling — helps families understand real risk, not just Punnett square ratios.	Does not account for new mutations or environmental effects on phenotype.
Can be combined with pedigree data across multiple generations for complex family analysis.	Requires accurate knowledge of parental genotypes; if parents' genotypes are uncertain, the calculation becomes more complex.
Works with any single-gene trait following Mendelian inheritance.	Polygenic traits (controlled by many genes) need different statistical tools.

⚠️ Common Mistake Alert!

Many students say the probability of being a carrier from a Aa × Aa cross is 1/2 (50%). That's the unconditional probability — before you know anything about the offspring's phenotype. Once you know the offspring is unaffected, you must recalculate. The correct answer is 2/3. Always ask yourself: "Am I given extra information? If yes, I need conditional probability."

✦ KEY TAKEAWAY

Imagine you're at a party with 4 friends: one wearing a red hat, two wearing blue hats, and one wearing a green hat. If someone tells you "your friend is NOT wearing a green hat," you now only consider 3 friends. The chance of picking the one in the red hat goes from 1/4 to 1/3. Conditional probability in genetics works exactly the same way — knowing the phenotype removes impossible genotypes and changes the odds.

SECTION 8

Connection to Advanced Genetics & Bayesian Analysis

The conditional probability skills you've learned here are the foundation for Bayesian analysis in genetics. Bayes' theorem is a more powerful version of the same idea — it lets you update probabilities as you gather more and more evidence, like test results or the phenotypes of additional family members.

Basic conditional probability vs. full Bayesian analysis

Feature	Basic Conditional Probability	Full Bayesian Analysis
Information used	One piece of evidence (e.g., phenotype of the individual)	Multiple pieces of evidence (e.g., phenotypes of parents, siblings, offspring, and test results)
Typical use	Simple single-cross problems (Aa × Aa)	Multi-generation pedigrees with incomplete information
Math complexity	Fractions and basic division	Requires organizing prior and posterior probabilities, sometimes in a Bayesian table
Example question	"Given this child is unaffected, what's the probability they are a carrier?"	"This unaffected woman has 3 unaffected sons. What is the updated probability she is a carrier of an X-linked disorder?"

In advanced genetics courses and real genetic counseling, Bayesian tables are used to handle situations where you have several generations of data. The good news? The core logic is identical to what you've learned today: identify what you know, eliminate impossibilities, and rescale the remaining probabilities so they add up to 1. Mastering the simple version prepares you perfectly for the advanced version.

SECTION 9

Practice Problems

PROBLEM 1 — CONCEPTUAL

In a cross between two heterozygous parents (Aa × Aa), the unconditional probability of an offspring being Aa is 2/4 = 1/2. Explain in your own words why the conditional probability of being Aa — given that the offspring shows the dominant phenotype — is different from 1/2.

PROBLEM 2 — BASIC CALCULATION

Two parents are both carriers for sickle cell trait (Ss × Ss). Their child does NOT have sickle cell disease. What is the probability the child is a carrier (Ss)? Use the conditional probability formula.

PROBLEM 3 — INTERMEDIATE

Both of Jamal's parents are carriers for PKU (Pp × Pp). Jamal is unaffected. He marries Priya, whose parents are also both carriers (Pp × Pp), and Priya is also unaffected. What is the probability that their first child will have PKU?

PROBLEM 4 — APPLIED

A genetic counselor tells a couple the following: the wife's brother has cystic fibrosis (CF), which means both of her parents must be carriers (Cc). The wife herself is healthy. The husband has no family history of CF, but in the general population, about 1 in 25 people are CF carriers. What is the probability their child will have CF?

PROBLEM 5 — CRITICAL THINKING

In a Bb × Bb cross, an offspring displays the dominant phenotype. You now do a test cross by mating this offspring with a homozygous recessive individual (bb). The test cross produces 5 offspring, all of which show the dominant phenotype. Given this new evidence, what is the updated probability that the original offspring is BB rather than Bb? (Hint: Consider how the test cross results change the odds.)

SUMMARY

Lesson Summary

Conditional probability is the mathematical tool for updating the chance of a genotype once you know an organism's phenotype. The core formula is P(A | B) = P(A and B) / P(B). In practice, this means you eliminate impossible outcomes from the Punnett square and recalculate probabilities using only the outcomes that remain.

The most common application: when both parents are carriers (Aa × Aa) and the offspring is unaffected, the probability the offspring is a carrier is 2/3, not 1/2. This result is critical in genetic counseling and appears in problems involving cystic fibrosis, sickle cell disease, PKU, and other autosomal recessive conditions. Mastering this concept prepares you for advanced Bayesian analysis used in multi-generation pedigrees.