Limits...
Linkage disequilibrium compared between five populations of domestic sheep.

Meadows JR, Chan EK, Kijas JW - BMC Genet. (2008)

Bottom Line: Sheep populations were selected which were inbred (Macarthur Merino), highly heterogeneous (Merino) or intermediate between these two extremes.This facilitated analysis and comparison of LD (x2') between populations.The strength and magnitude of LD was found to differ markedly between breeds and aligned closely with both observed levels of genetic diversity and expectations based on breed history.

View Article: PubMed Central - HTML - PubMed

Affiliation: CSIRO Livestock Industries, Level 5 Queensland Bioscience Precinct, 306 Carmody Road, St Lucia 4067, Australia. Jennifer.Meadows@csiro.au

ABSTRACT

Background: The success of genome-wide scans depends on the strength and magnitude of linkage disequilibrium (LD) present within the populations under investigation. High density SNP arrays are currently in development for the sheep genome, however little is known about the behaviour of LD in this livestock species. This study examined the behaviour of LD within five sheep populations using two LD metrics, D' and x2'. Four economically important Australian sheep flocks, three pure breeds (White Faced Suffolk, Poll Dorset, Merino) and a crossbred population (Merino x Border Leicester), along with an inbred Australian Merino museum flock were analysed.

Results: Short range LD (0 - 5 cM) was observed in all five populations, however the persistence with increasing distance and magnitude of LD varied considerably between populations. Average LD (x2') for markers spaced up to 20 cM exceeded the non-syntenic average within the White Faced Suffolk, Poll Dorset and Macarthur Merino. LD decayed faster within the Merino and Merino x Border Leicester, with LD below or consistent with observed background levels. Using marker-marker LD as a guide to the behaviour of marker-QTL LD, estimates of minimum marker spacing were made. For a 95% probability of detecting QTL, a microsatellite marker would be required every 0.1 - 2.5 centimorgans, depending on the population used.

Conclusion: Sheep populations were selected which were inbred (Macarthur Merino), highly heterogeneous (Merino) or intermediate between these two extremes. This facilitated analysis and comparison of LD (x2') between populations. The strength and magnitude of LD was found to differ markedly between breeds and aligned closely with both observed levels of genetic diversity and expectations based on breed history. This confirmed that breed specific information is likely to be important for genome wide selection and during the design of successful genome scans where tens of thousands of markers will be required.

Show MeSH

Related in: MedlinePlus

Linkage disequilibrium (x2') as a function of genetic distance. Each population is plotted in a separate panel. The absolute values of x2' (green circles) are plotted as a function of the genetic distance separating each marker pair (cM). Note the Y axis scale (x2') is not the same for each population. The mean value of x2' within defined distance bins is shown as horizontal green bars and contained within Table 2. The decay of LD modelled as a function of distance according to formula 3 is shown using black diamonds. Two significance thresholds are indicated using horizontal lines. The first represents the average x2' value obtained between non-syntenic marker pairs (orange line) while the second represents the 5% significance threshold (red line).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC2572059&req=5

Figure 2: Linkage disequilibrium (x2') as a function of genetic distance. Each population is plotted in a separate panel. The absolute values of x2' (green circles) are plotted as a function of the genetic distance separating each marker pair (cM). Note the Y axis scale (x2') is not the same for each population. The mean value of x2' within defined distance bins is shown as horizontal green bars and contained within Table 2. The decay of LD modelled as a function of distance according to formula 3 is shown using black diamonds. Two significance thresholds are indicated using horizontal lines. The first represents the average x2' value obtained between non-syntenic marker pairs (orange line) while the second represents the 5% significance threshold (red line).

Mentions: Linkage disequilibrium was estimated for all marker pairs using the metric x2', a standardised chi-square statistic suitable for use with multi-allelic markers [15]. The values of x2' derived from chromosome 18 marker pairs were plotted as a function of increasing genetic distance (Figure 2). Figure 2 shows x2' derived from syntenic marker pairs (green circles) exceeded the average derived from non-syntenic markers (orange line) for closely spaced markers in each of the five populations tested. For example, average LD for markers separated by less than 5 cM in WFS (x2' = 0.167 ± 0.076) was well above the average observed using non-syntenic markers in the same population (x2' = 0.099 ± 0.047; Figure 2, Table 2). Short range LD was observed in all five populations, however LD was observed to persist over larger chromosomal distances in some populations. Average LD for markers spaced up to 20 cM exceeded the non-syntenic average within the WFS, PD and EMAI populations (Table 2, Figure 2). When x2' was compared against the 5% threshold for significant LD (red line, Figure 2), many fewer marker pairs display both the magnitude and significance which exceeds the critical level. This was particularly evident in MER and MxB where less than 9% of marker pair combinations had x2' which exceeded the 5% threshold. The threshold limits applied here (0.05 – 0.15, Table 2) did not appear unrealistically high when compared to those applied in commercial chicken (x2' range 0.07 – 0.25) [16].


Linkage disequilibrium compared between five populations of domestic sheep.

Meadows JR, Chan EK, Kijas JW - BMC Genet. (2008)

Linkage disequilibrium (x2') as a function of genetic distance. Each population is plotted in a separate panel. The absolute values of x2' (green circles) are plotted as a function of the genetic distance separating each marker pair (cM). Note the Y axis scale (x2') is not the same for each population. The mean value of x2' within defined distance bins is shown as horizontal green bars and contained within Table 2. The decay of LD modelled as a function of distance according to formula 3 is shown using black diamonds. Two significance thresholds are indicated using horizontal lines. The first represents the average x2' value obtained between non-syntenic marker pairs (orange line) while the second represents the 5% significance threshold (red line).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC2572059&req=5

Figure 2: Linkage disequilibrium (x2') as a function of genetic distance. Each population is plotted in a separate panel. The absolute values of x2' (green circles) are plotted as a function of the genetic distance separating each marker pair (cM). Note the Y axis scale (x2') is not the same for each population. The mean value of x2' within defined distance bins is shown as horizontal green bars and contained within Table 2. The decay of LD modelled as a function of distance according to formula 3 is shown using black diamonds. Two significance thresholds are indicated using horizontal lines. The first represents the average x2' value obtained between non-syntenic marker pairs (orange line) while the second represents the 5% significance threshold (red line).
Mentions: Linkage disequilibrium was estimated for all marker pairs using the metric x2', a standardised chi-square statistic suitable for use with multi-allelic markers [15]. The values of x2' derived from chromosome 18 marker pairs were plotted as a function of increasing genetic distance (Figure 2). Figure 2 shows x2' derived from syntenic marker pairs (green circles) exceeded the average derived from non-syntenic markers (orange line) for closely spaced markers in each of the five populations tested. For example, average LD for markers separated by less than 5 cM in WFS (x2' = 0.167 ± 0.076) was well above the average observed using non-syntenic markers in the same population (x2' = 0.099 ± 0.047; Figure 2, Table 2). Short range LD was observed in all five populations, however LD was observed to persist over larger chromosomal distances in some populations. Average LD for markers spaced up to 20 cM exceeded the non-syntenic average within the WFS, PD and EMAI populations (Table 2, Figure 2). When x2' was compared against the 5% threshold for significant LD (red line, Figure 2), many fewer marker pairs display both the magnitude and significance which exceeds the critical level. This was particularly evident in MER and MxB where less than 9% of marker pair combinations had x2' which exceeded the 5% threshold. The threshold limits applied here (0.05 – 0.15, Table 2) did not appear unrealistically high when compared to those applied in commercial chicken (x2' range 0.07 – 0.25) [16].

Bottom Line: Sheep populations were selected which were inbred (Macarthur Merino), highly heterogeneous (Merino) or intermediate between these two extremes.This facilitated analysis and comparison of LD (x2') between populations.The strength and magnitude of LD was found to differ markedly between breeds and aligned closely with both observed levels of genetic diversity and expectations based on breed history.

View Article: PubMed Central - HTML - PubMed

Affiliation: CSIRO Livestock Industries, Level 5 Queensland Bioscience Precinct, 306 Carmody Road, St Lucia 4067, Australia. Jennifer.Meadows@csiro.au

ABSTRACT

Background: The success of genome-wide scans depends on the strength and magnitude of linkage disequilibrium (LD) present within the populations under investigation. High density SNP arrays are currently in development for the sheep genome, however little is known about the behaviour of LD in this livestock species. This study examined the behaviour of LD within five sheep populations using two LD metrics, D' and x2'. Four economically important Australian sheep flocks, three pure breeds (White Faced Suffolk, Poll Dorset, Merino) and a crossbred population (Merino x Border Leicester), along with an inbred Australian Merino museum flock were analysed.

Results: Short range LD (0 - 5 cM) was observed in all five populations, however the persistence with increasing distance and magnitude of LD varied considerably between populations. Average LD (x2') for markers spaced up to 20 cM exceeded the non-syntenic average within the White Faced Suffolk, Poll Dorset and Macarthur Merino. LD decayed faster within the Merino and Merino x Border Leicester, with LD below or consistent with observed background levels. Using marker-marker LD as a guide to the behaviour of marker-QTL LD, estimates of minimum marker spacing were made. For a 95% probability of detecting QTL, a microsatellite marker would be required every 0.1 - 2.5 centimorgans, depending on the population used.

Conclusion: Sheep populations were selected which were inbred (Macarthur Merino), highly heterogeneous (Merino) or intermediate between these two extremes. This facilitated analysis and comparison of LD (x2') between populations. The strength and magnitude of LD was found to differ markedly between breeds and aligned closely with both observed levels of genetic diversity and expectations based on breed history. This confirmed that breed specific information is likely to be important for genome wide selection and during the design of successful genome scans where tens of thousands of markers will be required.

Show MeSH
Related in: MedlinePlus