Biology Reference
In-Depth Information
graphic , meaning that ( h 1 , h 2 ,..., h n ) < ( h 1 , h 2 ,..., h n ) if the first compo-
nent with different values satisfies h i < h i . The proof of this lemma is simple
and omitted.
Lemma 6.1. If the elements of
are ordered lexicographically, then for unique
i and j between 1 and 2 n we have that h i + h i +1
H
= h j + h j +1 and that h j +
h (2 n −j +1) =(0 , 0 ,..., 0) .
Since haplotypes mate in unique pairs to form a genotype, the degree of every
node in
is even. An immediate consequence of this observation is that not ev-
ery bipartite graph supports diversity. Moreover, even if every node of a bipartite
graph has even degree it does not mean the graph supports diversity. As an exam-
ple, the complete bipartite graph K 2 , 2 does not support diversity because any η
and γ that satisfies the third and fourth conditions of the definition violates the fact
that γ is one-to-one. Theorem 6.1 does not characterize the graphs that support
diversity, but it does establish a necessary condition.
V
,η,γ ) be a diversity graph. Let w 1 and w 2 be in
Theorem 6.1. Let (
V
,
W
,
E
W
,
with w 1
= w 2 . Then,
deg( w 1 ) , deg( w 2 )
N ( w 1 )
N ( w 2 )
max
{
}≥
2
|
|
.
H = ( v ): v ∈ N ( w 1 ) ∩N ( w 2 ) }
Proof.
Let
and assume to the contrary that
deg( w 1 ) , deg( w 2 )
|H |
max
{
}
< 2
.
Since haplotypes mate in unique pairs, there must be fewer than
|H |
pairs of
haplotypes mating to form each of γ ( w 1 )= g
1 and γ ( w 2 )= g
2 . It follows that
1 , h
2 , h
3 and h
4 in
H
1 + h
2 = g
1 and h
3 + h
4 = g
2 .
there exists h
such that h
1
j =2, which means h j =1for all h ∈H . It follows that g
2
Suppose that g
j =2.
1
2
1
2
Similarly, if g
j =
2,wehavethat g
j =
2. Suppose that g
j =0.Then g
j can
1
not be 2 or
2 because if so, the same argument would guarantee that g
j is 2 or
2
2, respectively. We conclude that g
j =0.Since j was arbitrary, we have the
1 = g
2 .
contradiction that g
We now turn our direction to a matrix equation that is satisfied by every di-
versity graph. Consider the diversity graph (
V
,
W
,
E
,η,γ ),where
|V|
= m and
|W|
= k . List the elements of
V
as v 1 ,v 2 ,...,v m and
W
as w 1 ,w 2 ,...,w k ,
and assume that η ( v i )= h
i and γ ( w i )= g
i .Let H be the m
n matrix so that
H ( i,j ) = h i j ,andlet G be the k ×n matrix defined by G ( i,j ) = g i j .Alsolet E be
the m
×
k biadjacency matrix —i.e. E ( i,j ) =1if ( v i ,w j )
and E ( i,j ) =0
otherwise. Note that the column sums of E must be even from the definition of a
diversity graph.
×
∈E
Search WWH ::




Custom Search