Limits...
The confounding effect of population structure on Bayesian skyline plot inferences of demographic history.

Heller R, Chikhi L, Siegismund HR - PLoS ONE (2013)

Bottom Line: We found that a balanced sampling strategy whereby samples are distributed on several populations provides the best scheme for inferring demographic change over a typical time scale.Analyses of data from a structured African buffalo population demonstrate how BSP results can be strengthened by simulations.We recommend that sample selection should be carefully considered in relation to population structure previous to BSP analyses, and that alternative scenarios should be evaluated when interpreting signals of population size change.

View Article: PubMed Central - PubMed

Affiliation: Instituto Gulbenkian de Ciência, Oeiras, Portugal. rheller@bio.ku.dk

ABSTRACT
Many coalescent-based methods aiming to infer the demographic history of populations assume a single, isolated and panmictic population (i.e. a Wright-Fisher model). While this assumption may be reasonable under many conditions, several recent studies have shown that the results can be misleading when it is violated. Among the most widely applied demographic inference methods are Bayesian skyline plots (BSPs), which are used across a range of biological fields. Violations of the panmixia assumption are to be expected in many biological systems, but the consequences for skyline plot inferences have so far not been addressed and quantified. We simulated DNA sequence data under a variety of scenarios involving structured populations with variable levels of gene flow and analysed them using BSPs as implemented in the software package BEAST. Results revealed that BSPs can show false signals of population decline under biologically plausible combinations of population structure and sampling strategy, suggesting that the interpretation of several previous studies may need to be re-evaluated. We found that a balanced sampling strategy whereby samples are distributed on several populations provides the best scheme for inferring demographic change over a typical time scale. Analyses of data from a structured African buffalo population demonstrate how BSP results can be strengthened by simulations. We recommend that sample selection should be carefully considered in relation to population structure previous to BSP analyses, and that alternative scenarios should be evaluated when interpreting signals of population size change.

Show MeSH

Related in: MedlinePlus

Two demographic scenarios under a real data informed island model.As Figs. 1 and 2, but the island model was modified to conform to the migration matrix estimated for a real biological system, the African buffalo. Only pooled and scattered sampling is shown. Bold lines mark the appropriate simulated population size. Replication was slightly different from Fig. 1 and 2 because it now matters which demes were included in the sample, see main text. Note also that the number of demes was 34, so the sum of the deme size differs from Fig. 1 and 2 (17,000 as opposed to 20,000 females).
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC3646956&req=5

pone-0062992-g003: Two demographic scenarios under a real data informed island model.As Figs. 1 and 2, but the island model was modified to conform to the migration matrix estimated for a real biological system, the African buffalo. Only pooled and scattered sampling is shown. Bold lines mark the appropriate simulated population size. Replication was slightly different from Fig. 1 and 2 because it now matters which demes were included in the sample, see main text. Note also that the number of demes was 34, so the sum of the deme size differs from Fig. 1 and 2 (17,000 as opposed to 20,000 females).

Mentions: Our simulations mimicking the inferred gene flow connecting 34 African buffalo populations revealed that under this structural configuration, the risk of a false signal of population size change was relatively low (Fig. 3) and resembled those of the high-migration scenarios under the idealised island model (Fig. 1G–I). Interestingly, the local sampling–when analysed deme by deme–revealed that there was a high correlation between the deme connectedness (measured as the mean of all pairwise FST values involving a given deme) and PSC (Fig. S4). This shows that within a structured population with unequal deme connectedness, the risk of false positives of population decline depends on which demes are sampled. We expanded on this observation by simulating local sampling in a structured population with a wide range of gene flow among demes (Nfm 0.12–79.86, corresponding to an equilibrium FST of 0.007–0.827; see Supporting Information S1 for details). This revealed a clear separation between two phases in the relation between FST and PSC (which quantifies the risk of a structure effect): when Nfm <2 (FST <0.2), there was a strong positive correlation between the two and when Nfm >2, the correlation was negative (Fig. S4). The latter was initially surprising, but then we considered that in very isolated demes, there is a high probability that all lineages coalesce in the scattering phase (i.e. in the sampled deme) so that there will be no collecting phase (see the Discussion). We show EBSPs and PSC histograms for the two most extreme demes in terms of connectedness (Fig. S5) to illustrate the importance of deme connectedness.


The confounding effect of population structure on Bayesian skyline plot inferences of demographic history.

Heller R, Chikhi L, Siegismund HR - PLoS ONE (2013)

Two demographic scenarios under a real data informed island model.As Figs. 1 and 2, but the island model was modified to conform to the migration matrix estimated for a real biological system, the African buffalo. Only pooled and scattered sampling is shown. Bold lines mark the appropriate simulated population size. Replication was slightly different from Fig. 1 and 2 because it now matters which demes were included in the sample, see main text. Note also that the number of demes was 34, so the sum of the deme size differs from Fig. 1 and 2 (17,000 as opposed to 20,000 females).
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC3646956&req=5

pone-0062992-g003: Two demographic scenarios under a real data informed island model.As Figs. 1 and 2, but the island model was modified to conform to the migration matrix estimated for a real biological system, the African buffalo. Only pooled and scattered sampling is shown. Bold lines mark the appropriate simulated population size. Replication was slightly different from Fig. 1 and 2 because it now matters which demes were included in the sample, see main text. Note also that the number of demes was 34, so the sum of the deme size differs from Fig. 1 and 2 (17,000 as opposed to 20,000 females).
Mentions: Our simulations mimicking the inferred gene flow connecting 34 African buffalo populations revealed that under this structural configuration, the risk of a false signal of population size change was relatively low (Fig. 3) and resembled those of the high-migration scenarios under the idealised island model (Fig. 1G–I). Interestingly, the local sampling–when analysed deme by deme–revealed that there was a high correlation between the deme connectedness (measured as the mean of all pairwise FST values involving a given deme) and PSC (Fig. S4). This shows that within a structured population with unequal deme connectedness, the risk of false positives of population decline depends on which demes are sampled. We expanded on this observation by simulating local sampling in a structured population with a wide range of gene flow among demes (Nfm 0.12–79.86, corresponding to an equilibrium FST of 0.007–0.827; see Supporting Information S1 for details). This revealed a clear separation between two phases in the relation between FST and PSC (which quantifies the risk of a structure effect): when Nfm <2 (FST <0.2), there was a strong positive correlation between the two and when Nfm >2, the correlation was negative (Fig. S4). The latter was initially surprising, but then we considered that in very isolated demes, there is a high probability that all lineages coalesce in the scattering phase (i.e. in the sampled deme) so that there will be no collecting phase (see the Discussion). We show EBSPs and PSC histograms for the two most extreme demes in terms of connectedness (Fig. S5) to illustrate the importance of deme connectedness.

Bottom Line: We found that a balanced sampling strategy whereby samples are distributed on several populations provides the best scheme for inferring demographic change over a typical time scale.Analyses of data from a structured African buffalo population demonstrate how BSP results can be strengthened by simulations.We recommend that sample selection should be carefully considered in relation to population structure previous to BSP analyses, and that alternative scenarios should be evaluated when interpreting signals of population size change.

View Article: PubMed Central - PubMed

Affiliation: Instituto Gulbenkian de Ciência, Oeiras, Portugal. rheller@bio.ku.dk

ABSTRACT
Many coalescent-based methods aiming to infer the demographic history of populations assume a single, isolated and panmictic population (i.e. a Wright-Fisher model). While this assumption may be reasonable under many conditions, several recent studies have shown that the results can be misleading when it is violated. Among the most widely applied demographic inference methods are Bayesian skyline plots (BSPs), which are used across a range of biological fields. Violations of the panmixia assumption are to be expected in many biological systems, but the consequences for skyline plot inferences have so far not been addressed and quantified. We simulated DNA sequence data under a variety of scenarios involving structured populations with variable levels of gene flow and analysed them using BSPs as implemented in the software package BEAST. Results revealed that BSPs can show false signals of population decline under biologically plausible combinations of population structure and sampling strategy, suggesting that the interpretation of several previous studies may need to be re-evaluated. We found that a balanced sampling strategy whereby samples are distributed on several populations provides the best scheme for inferring demographic change over a typical time scale. Analyses of data from a structured African buffalo population demonstrate how BSP results can be strengthened by simulations. We recommend that sample selection should be carefully considered in relation to population structure previous to BSP analyses, and that alternative scenarios should be evaluated when interpreting signals of population size change.

Show MeSH
Related in: MedlinePlus