Limits...
Genetic crossovers are predicted accurately by the computed human recombination map.

Khil PP, Camerini-Otero RD - PLoS Genet. (2010)

Bottom Line: This instability and the reported high level of inter-individual variation in meiotic recombination puts in question the accuracy of the calculated hotspot map, which is based on the summation of past genetic crossovers.To estimate the accuracy of the computed recombination rate map, we have mapped genetic crossovers to a median resolution of 70 Kb in 10 CEPH pedigrees.An in-depth examination of not-predicted crossovers shows that they are preferentially located in regions where hotspots are found in other populations.

View Article: PubMed Central - PubMed

Affiliation: Genetics and Biochemistry Branch, The National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, Maryland, United States of America.

ABSTRACT
Hotspots of meiotic recombination can change rapidly over time. This instability and the reported high level of inter-individual variation in meiotic recombination puts in question the accuracy of the calculated hotspot map, which is based on the summation of past genetic crossovers. To estimate the accuracy of the computed recombination rate map, we have mapped genetic crossovers to a median resolution of 70 Kb in 10 CEPH pedigrees. We then compared the positions of crossovers with the hotspots computed from HapMap data and performed extensive computer simulations to compare the observed distributions of crossovers with the distributions expected from the calculated recombination rate maps. Here we show that a population-averaged hotspot map computed from linkage disequilibrium data predicts well present-day genetic crossovers. We find that computed hotspot maps accurately estimate both the strength and the position of meiotic hotspots. An in-depth examination of not-predicted crossovers shows that they are preferentially located in regions where hotspots are found in other populations. In summary, we find that by combining several computed population-specific maps we can capture the variation in individual hotspots to generate a hotspot map that can predict almost all present-day genetic crossovers.

Show MeSH

Related in: MedlinePlus

Most CEPH crossovers not predicted by CEU hotspots overlap hotspots from other populations.The fraction of crossover intervals not predicted by CEU hotspots that overlap hotspots found in YRI, CHB, JPT, or any of the other HapMap PhaseII populations (YRI, CHB, or JPT) is plotted. For comparison, the same fraction of crossovers overlapping hotspots from other populations (mean and 95% CI) is plotted for crossovers re-distributed according to the population-averaged map and randomly distributed crossovers.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC2813264&req=5

pgen-1000831-g005: Most CEPH crossovers not predicted by CEU hotspots overlap hotspots from other populations.The fraction of crossover intervals not predicted by CEU hotspots that overlap hotspots found in YRI, CHB, JPT, or any of the other HapMap PhaseII populations (YRI, CHB, or JPT) is plotted. For comparison, the same fraction of crossovers overlapping hotspots from other populations (mean and 95% CI) is plotted for crossovers re-distributed according to the population-averaged map and randomly distributed crossovers.

Mentions: The finding that the population-averaged map is in closer agreement with the distribution of crossovers compared to the CEU map suggests that hotspots from other populations may be in fact active in the CEU sample but not detected in the CEU profile. Thus, we asked where such not predicted crossovers are located relative to hotspots detected in other populations. We find that, depending on the accuracy of mapping, between 50 and 61 percent of not predicted crossovers overlap at least one hotspot from another population (YRI, CHB or JPT) (results for CEPH crossovers are shown in Figure 5 and for Hutterite crossovers are shown in Figure S15). Importantly, this proportion is significantly higher than expected if crossovers would be distributed randomly (P<0.001 by simulation, see Figure 5 and Figure S15 for all crossovers; data are not shown for other subsets of crossovers) meaning that crossovers are preferentially located in regions where hotspots are found in other populations. Furthermore, as one might expect, the fraction of not predicted crossovers that overlap at least one hotspot from another population is similar to the expected proportion if crossovers would be distributed according to the population-averaged map (Figure 5, Figure S15).


Genetic crossovers are predicted accurately by the computed human recombination map.

Khil PP, Camerini-Otero RD - PLoS Genet. (2010)

Most CEPH crossovers not predicted by CEU hotspots overlap hotspots from other populations.The fraction of crossover intervals not predicted by CEU hotspots that overlap hotspots found in YRI, CHB, JPT, or any of the other HapMap PhaseII populations (YRI, CHB, or JPT) is plotted. For comparison, the same fraction of crossovers overlapping hotspots from other populations (mean and 95% CI) is plotted for crossovers re-distributed according to the population-averaged map and randomly distributed crossovers.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC2813264&req=5

pgen-1000831-g005: Most CEPH crossovers not predicted by CEU hotspots overlap hotspots from other populations.The fraction of crossover intervals not predicted by CEU hotspots that overlap hotspots found in YRI, CHB, JPT, or any of the other HapMap PhaseII populations (YRI, CHB, or JPT) is plotted. For comparison, the same fraction of crossovers overlapping hotspots from other populations (mean and 95% CI) is plotted for crossovers re-distributed according to the population-averaged map and randomly distributed crossovers.
Mentions: The finding that the population-averaged map is in closer agreement with the distribution of crossovers compared to the CEU map suggests that hotspots from other populations may be in fact active in the CEU sample but not detected in the CEU profile. Thus, we asked where such not predicted crossovers are located relative to hotspots detected in other populations. We find that, depending on the accuracy of mapping, between 50 and 61 percent of not predicted crossovers overlap at least one hotspot from another population (YRI, CHB or JPT) (results for CEPH crossovers are shown in Figure 5 and for Hutterite crossovers are shown in Figure S15). Importantly, this proportion is significantly higher than expected if crossovers would be distributed randomly (P<0.001 by simulation, see Figure 5 and Figure S15 for all crossovers; data are not shown for other subsets of crossovers) meaning that crossovers are preferentially located in regions where hotspots are found in other populations. Furthermore, as one might expect, the fraction of not predicted crossovers that overlap at least one hotspot from another population is similar to the expected proportion if crossovers would be distributed according to the population-averaged map (Figure 5, Figure S15).

Bottom Line: This instability and the reported high level of inter-individual variation in meiotic recombination puts in question the accuracy of the calculated hotspot map, which is based on the summation of past genetic crossovers.To estimate the accuracy of the computed recombination rate map, we have mapped genetic crossovers to a median resolution of 70 Kb in 10 CEPH pedigrees.An in-depth examination of not-predicted crossovers shows that they are preferentially located in regions where hotspots are found in other populations.

View Article: PubMed Central - PubMed

Affiliation: Genetics and Biochemistry Branch, The National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, Maryland, United States of America.

ABSTRACT
Hotspots of meiotic recombination can change rapidly over time. This instability and the reported high level of inter-individual variation in meiotic recombination puts in question the accuracy of the calculated hotspot map, which is based on the summation of past genetic crossovers. To estimate the accuracy of the computed recombination rate map, we have mapped genetic crossovers to a median resolution of 70 Kb in 10 CEPH pedigrees. We then compared the positions of crossovers with the hotspots computed from HapMap data and performed extensive computer simulations to compare the observed distributions of crossovers with the distributions expected from the calculated recombination rate maps. Here we show that a population-averaged hotspot map computed from linkage disequilibrium data predicts well present-day genetic crossovers. We find that computed hotspot maps accurately estimate both the strength and the position of meiotic hotspots. An in-depth examination of not-predicted crossovers shows that they are preferentially located in regions where hotspots are found in other populations. In summary, we find that by combining several computed population-specific maps we can capture the variation in individual hotspots to generate a hotspot map that can predict almost all present-day genetic crossovers.

Show MeSH
Related in: MedlinePlus