Genetics by Computer: Quick Revision

A brief note on Genetics by Computer. Read it before a test. Try each question at the end before you open its answer.

One gene

• A gamete is one letter: list("Tt") gives ['T', 't'].
• A child is "".join(sorted(a + b)); the capital, dominant letter comes first.
• Two loops over the parents' letters fill the Punnett square.
• Counter(kids) counts genotypes. Tt with Tt gives 1 TT : 2 Tt : 1 tt, which is 3 tall to 1 short.
• Tt with tt, a test cross, gives 1 tall to 1 short.

Two genes

• product("Rr", "Yy") gives the four gametes RY, Ry, rY and ry.
• Four gametes meet four in 16 boxes. Sort each gene of the child on its own.
• RrYy with RrYy gives 9:3:3:1. Each gene alone still gives 3:1, and $\dfrac{3}{4} \times \dfrac{3}{4} = \dfrac{9}{16}$.
• A cross with rryy reveals a plant's hidden small letters.

Chance

• random.choice(parent) picks one gamete; one choice from each parent makes a child.
• random.seed(k) makes every run repeat the same numbers.
• A few plants wander far from the ratio; thousands settle near it.

A worked example

What does this print?

1. The boxes. W with w gives Ww twice; w with w gives ww twice.
2. The phenotypes. Ww holds a capital W, so it is round; ww is wrinkled.

It prints Counter({'round': 2, 'wrinkled': 2}), a 1:1 ratio.

Practice with new numbers

A program to try

The program crosses Pp with pp and then Pp with Pp, and counts the flower colours each time.

Mistakes to avoid

• No sort. Then tT and Tt are counted as two genotypes.
• Testing for TT only. One capital letter is enough for the dominant look.
• Two letters of one gene in a gamete. A gamete has one letter of each gene.
• Both random letters from one parent. Take one from each.
• Judging a ratio from a few plants. Chance needs many plants to settle.

Check yourself

1. Why is "".join(sorted(a + b)) used instead of a + b?

2. How many boxes does the square of AaBb with AaBb have, and why?

3. A round plant crossed with rr gives only round seeds. What is it?

4. A simulation of 40 plants gives 30 % short. Is the 3:1 rule wrong?