AGB 121: Principles of Animal Genetics and Population Genetics (2+1)

Methods of estimation

HERITABILITY ESTIMATION METHODS

• The degree of resemblance between offspring and parent is measured by regression coefficient and that between full or half sib is measured by correlation.
• Coefficients of the casual component (variance components) in the Covariance (COV) with two factor interactions of relatives

Relatives	Variance components and their contribution
	VA	VD	VAA	VAD	VDD
Offspring – parent : covop	1/2	-	1/4	-	-
Half sibs : cov(HS)	1/4	-	1/16	-	-
Full sibs : cov(FS)	1/2	1/4	1/4	1/8	1/16

• In case of full sib the COV includes V_Ec (common environmental circumstances that cause differences between unrelated individual are not a cause of differences between members of the same family).
• Therefore, by observing the phenotypic covariance of different sorts of relationship the amount V_A obtained is

Relatives	Amount of VA provided by COV
Offspring and one parent	½ VA
Offspring and mid-parent	½ VA
Half sib	¼ VA
Full sib	½VA+ ¼ VD + VEc

Heritability estimation methods

• Regression method Regression of offspring on one parent
  • Regression of offspring on mid parent
  • Intra sire regression of offspring on dam
• Correlation method:
  • Half sib correlation
  • Full sib correlation
• Using twin data in human [Dizygotic (fraternal) and Monozygotic (identical) twins]
  • The genetic covariance of twins is very simple. In case of Dizygotic twins the genetic COV are related as full sibs. The monozygotic twins have identical genotypes, so there is no genetic variance within pairs and the whole of the genetic variance appears in the between pair component. The genetic COV is therefore COV _(MZ): V_G
• Heritability can also be estimated from selection experiment data

i.e. R = h²S

h² = R/S

Where R = Response to selection and S = Selection differential

The heritability estimated by this way is called Realized heritability.