360
FRIEDHELM HILDEBRANDT AND HEYMUT OMRAN
genetic distance) L(8) is divided by the likelihood of no linkage L(8=0.5).
This is called the
odds ratio = L(8)/L(8 = 0.5) or
odds ratio =
odds that the 2 loci linked at a recombination fraction e
odds that the 2 loci are unlinked (9=0.5)
where L is a likelihood or odds.
In words: "The odds ratio is the likelihood that there is linkage at an
assumed 8 divided by the likelihood that there is no linkage."
The logarithm of the odds ratio is called the
LOD score Z =log [L(8)/L(8=0.5)]
The LOD score is evaluated at the recombination fraction 8 where it is
maximal. This is called the maximum LOD score Z. A maximum LOD score
>3 is considered significant linkage between two loci. This is equivalent to
odds of 1:1000 for linkage. Ifthe LOD score is <-2, this is equivalent to significant exclusion of linkage between two loci.
When calculating LOD scores the data can be expressed in the LOD score
curve (see Figure 6).
Exclusion mapping
In exclusion mapping one tries by consecutively examining many markers
to exclude large areas of the human genome from linkage with a disease
Fig. 8. LODOUT5, the LODVIEW graphical display for human Chromosome 5. The abscissa
represents a map of Chromosome 5 extending from the p terminus (left) to the q terminus
(right) divided into intervals of 5 cM(K) of sex-averaged genetic distance (xK) according to
Kosambi's map function. MS markers are positioned relative to their approximate location
within the 5 cM(K) grid. Information on MS markers is shown above their respective positions
in the following order: symbol, name, (location) heterozygosity index. The sex-averaged genetic distance to the next MS marker towards the q-terminus is given in cM(K).
At the position of each marker, LOD scores Z(9=0.05) are shown as stacked histograms, each
segment representing the LOD score of one family (see legend), where positive LOD scores are
stackedontop of each other above the zero line and negative LOD scores are stackedontop of
each other below the zero line. Thus, the histograms allow direct ascertainment of which
families give positive or negative LOD scores at 9=0.05. Early-onset families are shown in
hatched segments (fam 1 - 6}. Late-onset families are shown in shades of gray (fam 7 -
11). In the position of each marker that generated a histogram, funnel-shaped curves surrounding each histogram in a symmetrical fashion show tracings of LOD scores at x(K) =
5, 10, 15, 20 cM(K} etc., on both sides of the marker position. A thick dashed line is drawn
at Z(9) = - 2. Its intersections with the two sides of a LOD score curve delimits the regions of
exclusion assuming the absence of non-allelic heterogeneity in the families examined. Thus,
graphical assessment of the width of an exclusion area is facilitated. The LOD score curve for
multipoint linkage is denoted by ( • ).
FRIEDHELM HILDEBRANDT AND HEYMUT OMRAN
genetic distance) L(8) is divided by the likelihood of no linkage L(8=0.5).
This is called the
odds ratio = L(8)/L(8 = 0.5) or
odds ratio =
odds that the 2 loci linked at a recombination fraction e
odds that the 2 loci are unlinked (9=0.5)
where L is a likelihood or odds.
In words: "The odds ratio is the likelihood that there is linkage at an
assumed 8 divided by the likelihood that there is no linkage."
The logarithm of the odds ratio is called the
LOD score Z =log [L(8)/L(8=0.5)]
The LOD score is evaluated at the recombination fraction 8 where it is
maximal. This is called the maximum LOD score Z. A maximum LOD score
>3 is considered significant linkage between two loci. This is equivalent to
odds of 1:1000 for linkage. Ifthe LOD score is <-2, this is equivalent to significant exclusion of linkage between two loci.
When calculating LOD scores the data can be expressed in the LOD score
curve (see Figure 6).
Exclusion mapping
In exclusion mapping one tries by consecutively examining many markers
to exclude large areas of the human genome from linkage with a disease
Fig. 8. LODOUT5, the LODVIEW graphical display for human Chromosome 5. The abscissa
represents a map of Chromosome 5 extending from the p terminus (left) to the q terminus
(right) divided into intervals of 5 cM(K) of sex-averaged genetic distance (xK) according to
Kosambi's map function. MS markers are positioned relative to their approximate location
within the 5 cM(K) grid. Information on MS markers is shown above their respective positions
in the following order: symbol, name, (location) heterozygosity index. The sex-averaged genetic distance to the next MS marker towards the q-terminus is given in cM(K).
At the position of each marker, LOD scores Z(9=0.05) are shown as stacked histograms, each
segment representing the LOD score of one family (see legend), where positive LOD scores are
stackedontop of each other above the zero line and negative LOD scores are stackedontop of
each other below the zero line. Thus, the histograms allow direct ascertainment of which
families give positive or negative LOD scores at 9=0.05. Early-onset families are shown in
hatched segments (fam 1 - 6}. Late-onset families are shown in shades of gray (fam 7 -
11). In the position of each marker that generated a histogram, funnel-shaped curves surrounding each histogram in a symmetrical fashion show tracings of LOD scores at x(K) =
5, 10, 15, 20 cM(K} etc., on both sides of the marker position. A thick dashed line is drawn
at Z(9) = - 2. Its intersections with the two sides of a LOD score curve delimits the regions of
exclusion assuming the absence of non-allelic heterogeneity in the families examined. Thus,
graphical assessment of the width of an exclusion area is facilitated. The LOD score curve for
multipoint linkage is denoted by ( • ).