Upload
others
View
1
Download
0
Embed Size (px)
Citation preview
Coefficients of Identity and Coancestry
Coefficients of Identity and Coancestry
Coefficients of Identity and Coancestry
Resemblance of Relative Pairs
I In our previous lectures we found that the phenotypic varianceof a trait can theoretically be partitioned into a number ofgenetic and environmental components.
I How these components can be estimated in practice is ofsignificant practical importance.
I A key principal for this is that genetic and environmentalsources of variance contribute differentially to the resemblancebetween different types of relatives.
I In fact, much of population and quantitative genetics theoryrests on the comparison of pairs of individuals - either on theirgenotypes or on their trait values, or both.
Coefficients of Identity and Coancestry
Resemblance of Relative Pairs
I Understanding the genetic covariance or correlation betweenrelatives is essential in the determination of heritability oftraits, as resemblance between relatives is a naturalconsequence of relatives inheriting copies of the same alleles.
I As is the case of the genetic variance, the genetic covariancebetween relatives can also be partitioned into componentsattributable to additive, dominance, and various epistaticeffects.
I Each of these variance component terms for thedecomposition of genetic covariances of a particular relativepair type are weighted by a coefficient that is calculated fromthe joint distribution of the alleles shared by the pair.
I Today we will focus on defining and calculating coefficients ofrelatedness for relative pairs.
Coefficients of Identity and Coancestry
Measures of RelatednessI A number of relatedness measures have been developed for
population genetics.I We will discuss some of the measures that play a central role
in quantitative-genetic formulations.I Measures of relatedness generally share two essential features.I First, relatedness can only be defined relative to some specific
frame of reference.I All members of a species or population share a common
ancestor at some point in time in a phylogeny.I We will generally consider recent ancestry as opposed to
ancient common ancestry.I For example, if no further genealogical information is observed
for the top generation of a pedigree, these individuals will beconsidered as the base and will be treated as unrelatedindividuals in the population.
Coefficients of Identity and Coancestry
Measures of Relatedness
I Second, measures of relatedness are based on the concept ofidentity by descent (IBD).
I Alleles that are IBD are exact copies of an ancestral allele.
I The distinction between identical by state (IBS) and IBD iscrucial.
I Alleles that have identical nucleotide sequences but havedescended from different ancestors in the reference populationare IBS but not IBD
I On the other hand, alleles that are IBD are necessarily IBSprovided there is no mutation of the inherited allele.
Coefficients of Identity and Coancestry
Pedigree IBD Sharing Example 1
I Consider the two pedigrees below. Which alleles are sharedIBD for the siblings in the two pedigrees?
I For the pedigree on the left, the A1 allele is shared IBD by thesiblings. For the pedigree on the right, with probability 1
2 theA1 allele is shared IBD by the siblings.
Coefficients of Identity and Coancestry
Pedigree IBD Sharing Example 2
I For the following nuclear family, there are six pairs of siblings.For each pair, how many alleles are shared IBD?
Coefficients of Identity and Coancestry
Identity States
I The IBD status of a single locus for an individual can berepresented by F and 1 − F , where F is the inbreedingcoefficient):
I A complete description of the IBD status among the fouralleles alleles carried by two individuals at a locus requires 15identity states
Coefficients of Identity and Coancestry
Identity States
I Consider a single locus in two individuals. There are fouralleles involved. We can describe IBD status using thefollowing schematic. Start with a grid of four nodes:
I A pair of nodes is connected with a line if the alleles are IBD.There are 15 possible configurations of IBD (next slide). Wedo not usually care about the distinction between maternaland paternal alleles, so the 15 possible configurations of IBDreduce to nine identity states.
Coefficients of Identity and Coancestry
Jacquard’s (1970) 9 Condensed Coefficients ofIdentity
1∆
∆6
8∆
∆7
∆9
∆2
∆3
∆4
∆5
Probability Identity state
A’s alleles
B’s alleles
Mat
ernal
Pat
ernal
Coefficients of Identity and Coancestry
Jacquard’s (1970) 9 Condensed Coefficients ofIdentity
I ∆i is the probability the two individuals have IBD status ofstate i .
I For a given pair of individuals, ∆1 through ∆9 are theprobabilities associated with each of the nine states.
I The first six states only occur if there is inbreeding, and inlarge, random-mating outbred populations, the probability ofthe first six states is essentially 0.
I That is, ∆1 through ∆6 have values of 0, and ∆7, ∆8, and∆9 are the probabilities that are of interest for relative pairs inoutbred populations.
I Note that ∆7 can be characterized as the probability that twoindividuals share both of their alleles IBD. ∆7 is also calledthe coefficient of fraternity.
Coefficients of Identity and Coancestry
Coefficients of Coancestry
I Now consider two alleles at a locus that are drawn at randomfrom two individuals i and j
I The probability that the two alleles are identical by descent(IBD) is defined to be the coefficient of coancestry orkinship coefficient and is often represented as θij .
I θij can be written in terms of the condensed coefficients ofidentity.
I Take a moment and write the kinship coefficient in terms ofthe condensed coefficients of identity
Coefficients of Identity and Coancestry
Coefficients of Coancestry
θij = ∆1 +1
2(∆3 + ∆5 + ∆7) +
1
4∆8
Coefficients of Identity and Coancestry
Path Counting
I Algorithms to obtain kinship coefficients often use a techniquecalled ”path counting”
I To get the kinship coefficient for I ’s parents, we would identifythe path linking the parents of I to their common ancestor(s).
Coefficients of Identity and Coancestry
Path Counting
I If the parents X ,Y of an individual I have ancestor A incommon, and if there are n individuals (including X ,Y , I ) inthe path linking the parents through A, then the inbreedingcoefficient of I , or the coancestry of X and Y , is
FI = θXY =
(1
2
)n
(1 + FA)
I If there are several ancestors, this expression is summed overall the ancestors.
Coefficients of Identity and Coancestry
Path Counting: Parent-Offspring
I The common ancestor of parent X and child Y is X . Thepath linking X ,Y to their common ancestor is YX and thishas n = 2 individuals. Therefore
θXY =
(1
2
)2
=1
4
Coefficients of Identity and Coancestry
Path Counting: Grandparent-Grandchild
I The common ancestor of grandparent X and grandchild Y isX . The path linking X ,Y to their common ancestor is YVXand this has n = 3 individuals. Therefore
θXY =
(1
2
)3
=1
8
Coefficients of Identity and Coancestry
Path Counting: Half Sibs
I The common ancestor of half sibs X and Y is V . The pathlinking X ,Y to their common ancestor is XVY and this hasn = 3 individuals. Therefore
θXY =
(1
2
)3
=1
8
Coefficients of Identity and Coancestry
Path Counting: Full Sibs
I Calculate the kinship coefficient for full sibs using pathcounting
Coefficients of Identity and Coancestry
Path Counting: Full Sibs
I The common ancestors of full sibs X and Y are U and V .The paths linking X and Y to their common ancestors areXUY and XVY and these each have n = 3 individuals.Therefore
θXY =
(1
2
)3
+
(1
2
)3
=1
4
Coefficients of Identity and Coancestry
Path Counting: First Cousins
I Calculate the kinship coefficient for first cousins using pathcounting
Coefficients of Identity and Coancestry
Path Counting: Full First Cousins
I The common ancestors of first cousins X and Y are U and V. The paths linking X and Y to their common ancestors areXCUDY and XCVDY and these each have n = 5 individuals.Therefore
θXY =
(1
2
)5
+
(1
2
)5
=1
16
Coefficients of Identity and Coancestry
Coefficients of RelatednessI Coefficients of Identity sharing probabilities and kinship
coefficients for a few outbred relative pairs
IBD Sharing Probabilites for OutbredsRelationship ∆7 ∆8 ∆9 θ
Parent-Offspring 0 1 0 14
Full Siblings 14
12
14
14
Half Siblings 0 12
12
18
Uncle-Nephew 0 12
12
18
First Cousins 0 14
34
116
Double First Cousins 116
616
916
18
Second Cousins 0 116
1516
164
Unrelated 0 0 1 0
I Note that for outbred populations ∆7 + ∆8 + ∆9 = 1I ∆7, ∆8, and ∆9 are often called IBD sharing probabilities
Coefficients of Identity and Coancestry
9 Coefficients of Identity: Pedigree Example
ss
Sib1 Sib2
I The pedigree above contains two siblings from parents whoare first cousins. Calculate the 9 condensed coefficients ofidentity for the siblings.
Coefficients of Identity and Coancestry
9 Coefficients of Identity: Pedigree Example
I For ∆1 we must have that the parents of the sibs share 1allele IBD and each parent passes down the IBD allele. Theparents share 1 allele IBD with probability 1
4 , so
∆1 =
(1
4
)(1
2
)4
=
(1
4
)(1
16
)=
1
64
I What is ∆2? ∆2 = 0 since it is not possible for both siblingsto be inbreed at the locus but with the alleles in sibling 1 tonot be IBD with the alleles of sibling 2 at the locus.
I Homework 3 will include the calculation of all 9 coefficients ofidentity as well as the kinship coefficients and inbreedingcoefficients for the siblings in this pedigree.