Apply the product and sum rules — Biology notes

WHY do we need these rules?

A Punnett square works fine for one gene (4 boxes) or even two genes (16 boxes). But three genes need 64 boxes, four genes need 256. Drawing those is slow and error-prone.

The deep insight: in a dihybrid (or trihybrid) cross, genes on different chromosomes assort independently (Mendel's Second Law). Independence is exactly the condition that lets us multiply probabilities. So instead of one huge square, we treat each gene as its own tiny problem and combine the answers.

WHAT are the two rules?

HOW do we derive them from first principles?

Think of probability as fraction of equally likely outcomes.

Product rule. Roll a fair gamete-maker twice. Event $A$ has probability $p$ , meaning it occurs in a fraction $p$ of trials. Among those trials, event $B$ (independent) occurs in fraction $q$ . So the fraction where both occur is "fraction of a fraction": $\underbrace{p}_{\text{A happens}}\times\underbrace{q}_{\text{then B happens}} = pq.$ Why multiply? Because "B given A" doesn't change ( $P(B\mid A)=P(B)=q$ by independence), so $P(A\cap B)=P(A)\,P(B\mid A)=pq$ .

Sum rule. Suppose genotype $Aa$ can be produced as either "egg $A$ , sperm $a$ " or "egg $a$ , sperm $A$ ." These are mutually exclusive (a single fertilization is one or the other, never both). Count outcomes: the favorable count is (count of route 1) + (count of route 2), so dividing by total outcomes: $P(A\cup B)=\frac{n_A+n_B}{N}=\frac{n_A}{N}+\frac{n_B}{N}=P(A)+P(B).$ Why add? Because we are pooling separate baskets of favorable outcomes with no overlap to double-count.

Figure — Apply the product and sum rules

Worked examples

Worked example Example 2 — Mixing both rules

Same cross $AaBbCc\times AaBbCc$ . What is $P(\text{dominant A, recessive bb, dominant C})$ , i.e. phenotype $A\_\,bb\,C\_$ ?

Step 1 (sum within a gene). $P(A\_)=P(AA)+P(Aa)=\tfrac14+\tfrac12=\tfrac34$ . Why add? $AA$ or $Aa$ both give the dominant phenotype, and they're mutually exclusive genotypes → sum rule. Similarly $P(C\_)=\tfrac34$ , and $P(bb)=\tfrac14$ .

Step 2 (product across genes). $P=\tfrac34\times\tfrac14\times\tfrac34=\tfrac{9}{64}.$ Why multiply? The three gene-outcomes are independent and must all hold (AND).

Forecast-then-Verify

Recall Predict before reading:

$AaBb \times AaBb$ , what is $P(\text{A\_B\_})$ (both dominant)? Forecast it, then check: $P(A\_)=\tfrac34$ , $P(B\_)=\tfrac34$ , independent → $\tfrac34\times\tfrac34=\boxed{\tfrac{9}{16}}$ . This is the "9" in the classic 9:3:3:1 ratio — proof the rules reproduce the Punnett result.

Common mistakes (Steel-manned)

Flashcards

When do you MULTIPLY probabilities in genetics?

When both independent events must occur together (AND); product rule

P(A)P(B)

When do you ADD probabilities in genetics?

When events are mutually exclusive and you want any one of them (OR); sum rule

P(A)+P(B)

What condition must hold to use the product rule?

The events must be independent (one outcome doesn't change the other's probability).

What condition must hold to use the sum rule?

The events must be mutually exclusive (cannot both happen in one trial).

P(aabbcc)

from

AaBbCc \times AaBbCc

\tfrac14\times\tfrac14\times\tfrac14=\tfrac{1}{64}

P(A\_)

from

Aa \times Aa

P(AA)+P(Aa)=\tfrac14+\tfrac12=\tfrac34

(sum rule).

How do you compute "at least one"?

1 - P(\text{none})

, where

P(\text{none})

uses the product rule.

Why can genes be treated separately in a dihybrid cross?

Independent assortment (Mendel's 2nd law) makes them independent, enabling the product rule.

Recall Feynman: explain to a 12-year-old

Imagine flipping two coins. To get heads AND heads, you need both to cooperate — that's rarer, so you multiply the chances ( $\tfrac12\times\tfrac12=\tfrac14$ ). To get heads OR tails on one coin, that's "either way works," so you add and it's easy ( $\tfrac12+\tfrac12=1$ , always!). Baby genes work the same: "this gene AND that gene" → multiply; "this version OR that version of the same gene" → add. That's the whole trick.

Connections

Law of Independent Assortment — why the product rule is valid across genes.
Punnett Square Method — the slow visual the rules replace.
Dihybrid Cross 9-3-3-1 Ratio — emerges directly from $\tfrac34\times\tfrac34$ etc.
Linked Genes and Recombination — where independence (and naive multiplication) breaks.
Probability Basics — the math foundation under all of this.

Concept Map

Hinglish (regional understanding)

Intuition Hinglish mein samjho

Dekho, genetics ke crosses asal mein probability ke problems hain. Do simple rule yaad rakho: product rule (AND) aur sum rule (OR). Jab do alag-alag genes ke outcome ek saath chahiye — jaise "aa AND bb" — tab probabilities ko multiply karte ho, kyunki ye independent events hain (Mendel ka independent assortment isi liye important hai). Jab ek hi gene ke andar do alag genotype same phenotype dete hain — jaise "AA OR Aa dono dominant dikhte hain" — tab probabilities ko add karte ho, kyunki ye mutually exclusive hain.

Ye rules itne kaam ke kyun hain? Kyunki trihybrid cross ka Punnett square 64 boxes ka hota hai — banana mushkil aur galti hone ke chance zyada. In rules se tum har gene ko alag chhota problem maan ke seedha answer nikaal lete ho. Jaise $AaBbCc \times AaBbCc$ mein $P(aabbcc) = \tfrac14 \times \tfrac14 \times \tfrac14 = \tfrac{1}{64}$ — bina koi square banaye!

Ek easy trick: AND ka matlab multiply, OR ka matlab add. Aur "at least one" type sawaal mein seedha complement use karo — $1 - P(\text{none})$ . Bas ek dhyaan rakho: multiply tabhi karo jab events sach mein independent hon (linked genes mein ye toot jaata hai), aur add tabhi karo jab events ek saath ho hi nahi sakte. Word padho — "and" ya "or" — aur rule khud decide ho jayega.

Test yourself — Mendelian Genetics

Connections