Gene Mapping with a Three-Point Cross

• the order of the gene loci
• the distance between them (in centiMorgans, cM)

• one parent is heterozygous for three linked alleles (C,c,Sh,sh,Bz,bz)
• and the other is homozygous for the recessive version of all three alleles (c,c,sh,sh,bz,bz)

• the gene order and
• gives a more accurate measurement of the distance in cM separating the three loci.

• Eight different phenotypes - representing the 8 possible genotypes (2³ = 8) are produced.
• The percentage of recombinants between C and Sh is 3.0%: 28/1000 of single recombinants plus 2/1000 double recombinants.

  (But note that if you had simply been doing a testcross of a C,c,Sh,sh dihybrid, you would not have seen the double recombinants - they restored the parental linkage - so you would have understated the distance at 2.8 cM.)

• That between Sh and Bz is 2.0%: 18/1000 single recombinants plus 2/1000 double recombinants.

  (Again, a testcross between a Sh,sh,Bz,bz dihybrid would have only shown the 1.8% recombinants.)

• That between C and Bz is 5.0%: 28 + 18 + 2 (2) = 50/1000.

  And, in this case, a C,c,Bz,bz dihybrid would have missed the two recombination events because they restored the parental configuration, so the "true" value between the loci is the sum of the single recombinants plus twice the frequency of double recombinants.

Compare these results with the (less accurate) mapping results for the same loci done by test crossing dihybrids rather than trihybrids.

A three-point cross also gives the gene order immediately.

1. Determine the rarest classes (here, C,sh,Bz and c,Sh,bz) because two crossovers between a pair of loci will be rarer than one.
2. The gene that is not found in the parental linkage occupies the middle locus. That may seem self-evident here because I placed the loci in what turned out to be the correct order. But if I had arranged them differently, e.g., C,Bz,sh and c,bz,Sh, it would not have mattered: the rule still holds: Sh is the middle locus.