Differences In Phenotypic Variance Research Articles

Differences in phenotypic variance between old and young congeneric species on a small island

AbstractAimInsular environments theoretically promote within‐species phenotypic variance, in part but not only because insular species enjoy a lower number of trophic interactions than their mainland counterparts. However, environmental factors apparently confound empirical comparisons between insular and mainland systems. To address this issue, we studied sympatric, congeneric species from the same small island, but that stemmed from separate colonization events. We predicted that the time since the species had colonized the island would correlate positively with the within‐species variance in morphometrics.LocationLifu island, New Caledonia, South Pacific.TaxaZosterops inornatus, Z. minutus, Z. lateralis (Passeriformes).MethodsWe measured seven morphological traits and used principal component analysis to compute axes of variation.ResultsThe within‐species variance in bill size and in feet/tail ratio corrected for body size increased with the estimated time since the species had been on the island. A similar but non‐significant trend existed for body size and bill width.Main ConclusionsThe morphometrics of the three species of Zosterops birds from Lifu supported the hypothesis that insular environments promote within‐species phenotypic variance, and cast a new light on the role of competition in the erosion of phenotypic variance.

Differences in the Phenotypic Mean and Variance Between Two Geographically Separated Populations of Wood Frog (Lithobates sylvaticus)

Intraspecific phenotypic variation between populations separated by large geographic distances is common. Differences in the mean and variance of traits among populations can be used to infer the relative strength, direction, and type of selection on traits. Patterns in the mean provide information on the type of selection, and patterns in variance provide information on the strength of selection. However, interpretation of mean/variance patterns is difficult when two traits are linked and strongly correlated to fitness because it is unlikely that each trait will reach phenotypic optima. In amphibians time to metamorphosis and size at metamorphosis are positively related both phenotypically and genetically. Using a common-garden experiment we investigated whether selection favours shorter time to metamorphosis or increased mass at metamorphosis between two populations which differ in the length of the post-metamorphic growing season by 2–4 weeks. Animals from the population a shorter growing season took longer to reach and metamorphosed at a greater mass, while animals from the population with a longer period for post metamorphic growth reached metamorphosis faster, but at a smaller mass. Greater phenotypic variance was observed in both traits in the population with the shorter growing season. These data suggest that animals from the population with a restricted growth period maximise mass at metamorphosis at the expense of longer larval periods while animals from population with the longer post-metamorphic growth period sacrifice mass at metamorphosis to shorten the larval period and maximise larval survival. Differences in phenotypic variance among populations suggest either directional or diversifying selection has acted on both traits.

Genomic Selection in Plant Breeding: A Comparison of Models

ABSTRACTSimulation and empirical studies of genomic selection (GS) show accuracies sufficient to generate rapid genetic gains. However, with the increased popularity of GS approaches, numerous models have been proposed and no comparative analysis is available to identify the most promising ones. Using eight wheat (Triticum aestivum L.), barley (Hordeum vulgare L.), Arabidopsis thaliana (L.) Heynh., and maize (Zea mays L.) datasets, the predictive ability of currently available GS models along with several machine learning methods was evaluated by comparing accuracies, the genomic estimated breeding values (GEBVs), and the marker effects for each model. While a similar level of accuracy was observed for many models, the level of overfitting varied widely as did the computation time and the distribution of marker effect estimates. Our comparisons suggested that GS in plant breeding programs could be based on a reduced set of models such as the Bayesian Lasso, weighted Bayesian shrinkage regression (wBSR, a fast version of BayesB), and random forest (RF) (a machine learning method that could capture nonadditive effects). Linear combinations of different models were tested as well as bagging and boosting methods, but they did not improve accuracy. This study also showed large differences in accuracy between subpopulations within a dataset that could not always be explained by differences in phenotypic variance and size. The broad diversity of empirical datasets tested here adds evidence that GS could increase genetic gain per unit of time and cost.

Sex differences in variance of intelligence across childhood

Why are males over-represented at the upper extremes of intelligence? One possibility for which there is some empirical support is that variance is greater among adult males. There is little published evidence of the development of that variability – is it manifest in early childhood or does it develop later? We explored sex differences in phenotypic variance in scores on a general ability factor extracted from several tests of verbal and non-verbal ability at ages 2, 3, 4, 7, 9 and 10 ( Ns from > 10,000 to > 2000) in a sample of British children. We found greater variance, by Levene’s test of homogeneity of variance, among boys at every age except age two despite the girls’ mean advantage from ages two to seven. Girls are significantly over-represented, as measured by chi-square tests, at the high tail and boys at the low tail at ages 2, 3 and 4. By age 10 the boys have a higher mean, greater variance and are over-represented in the high tail. Sex differences in variance emerge early – even before pre-school – suggesting that they are not determined by educational influences.

A Comparison of Twin Birthweight Data From Australia, the Netherlands, the United States, Japan, and South Korea: Are Genetic and Environmental Variations in Birthweight Similar in Caucasians and East Asians?

Birthweight has implications for physical and mental health in later life. Using data from Caucasian twins collected in Australia, the Netherlands and the United States, and from East Asian twins collected in Japan and South Korea, we compared the total phenotypic, genetic and environmental variances of birthweight between Caucasians and East Asians. Model-fitting analyses yielded four major findings. First, for both males and females, the total phenotypic variances of birthweight were about 45% larger in Caucasians than in East Asians. The larger phenotypic variances were mainly attributable to a greater shared environmental variance of birthweight in Caucasians (ranging from 62% to 67% of variance) than Asians (48% to 53%). Second, the genetic variance of birthweight was equal in Caucasians and East Asians for both males and females, explaining a maximum of 17% of variance. Third, small variations in total phenotypic variances of birthweight within Caucasians and within East Asians were mainly due to differences in nonshared environmental variances. We speculate that maternal effects (both genetic and environmental) explain the large shared environmental variance in birthweight and may account for the differences in phenotypic variance in birthweight between Caucasians and East Asians. Recent molecular findings and specific environmental factors that are subsumed by maternal effects are discussed.

A Comparison of Twin Birthweight Data From Australia, the Netherlands, the United States, Japan, and South Korea: Are Genetic and Environmental Variations in Birthweight Similar in Caucasians and East Asians?

AbstractBirthweight has implications for physical and mental health in later life. Using data from Caucasian twins collected in Australia, the Netherlands and the United States, and from East Asian twins collected in Japan and South Korea, we compared the total phenotypic, genetic and environmental variances of birthweight between Caucasians and East Asians. Model-fitting analyses yielded four major findings. First, for both males and females, the total phenotypic variances of birthweight were about 45% larger in Caucasians than in East Asians. The larger phenotypic variances were mainly attributable to a greater shared environmental variance of birth- weight in Caucasians (ranging from 62% to 67% of variance) than Asians (48% to 53%). Second, the genetic variance of birthweight was equal in Caucasians and East Asians for both males and females, explaining a maximum of 17% of variance. Third, small variations in total phenotypic variances of birthweight within Caucasians and within East Asians were mainly due to differences in nonshared environmental variances. We speculate that maternal effects (both genetic and environmental) explain the large shared environmental variance in birthweight and may account for the differences in phenotypic variance in birthweight between Caucasians and East Asians. Recent molecular findings and specific environmental factors that are subsumed by maternal effects are discussed.

ADHD: sibling interaction or dominance: an evaluation of statistical power.

Sibling interaction effects are suggested by a difference in phenotypic variance between monozygotic (MZ) twins and dizygotic (DZ) twins, and a pattern of twin correlations that is inconsistent with additive genetic influences. Notably, negative sibling interaction will result in MZ correlations which are more than twice as high as DZ correlations, a pattern also seen in the presence of genetic dominance. Negative sibling interaction effects have been reported in most genetic studies on Attention Deficit Hyperactivity Disorder (ADHD) and related phenotypes, while the presence of genetic dominance is not always considered in these studies. In the present paper the statistical power to detect both negative sibling interaction effects and genetic dominance is explored. Power calculations are presented for univariate models including sources of variation due to additive genetic influences, unique environmental influences, dominant genetic influences and a negative sibling interaction (i.e., contrast effect) between phenotypes of twins. Parameter values for heritability and contrast effects are chosen in accordance with published behavior genetic studies on ADHD and associated phenotypes. Results show that when both genetic dominance and contrast effects are truly present and using a classical twin design, genetic dominance is more likely to go undetected than the contrast effect. Failure to detect the presence of genetic dominance consequently gives rise to slightly biased estimates of additive genetic effects, unique environmental effects, and the contrast effect. Contrast effects are more easily detected in the absence of genetic dominance. If the significance of the contrast effect is evaluated while also including genetic dominance, small contrast effects are likely to go undetected, resulting in a relatively large bias in estimates of the other parameters. Alternative genetic designs, such as adding pairs of unrelated siblings reared together to a classical twin design, or adding non-twin siblings to twin pairs, greatly enhances the statistical power to detect contrast effects as well as the power to distinguish between genetic dominance and contrast effects.

Intraspecific hybridization, developmental stability and fitness in Drosophila mercatorum.

One of the possible effects of intraspecific hybridization is outbreeding depression, due to a breakdown of coadapted gene complexes, which can lead to reduced fitness and decreased developmental stability in hybrids. Alternatively, increased fitness and increased developmental stability in hybrids (hybrid vigour) may be a result of hybridization, probably due to increased heterozygosity. Developmental stability is assumed to be correlated with fitness and is commonly measured as fluctuating asymmetry or phenotypic variance. Drosophila mercatorum is capable of reproducing sexually, but also parthenogenetically in the laboratory. When selecting for parthenogenesis, the flies become homozygous in one generation; strong selection, therefore, is acting on the genome of these flies for coadaptation among genes. Intraspecific hybridization is therefore expected to have an impact when coadaptation is disrupted. Intraspecific hybridization between a parthenogenetic and a sexually reproducing strain of Drosophila mercatorum resulted in significant changes in fecundity as well as fluctuating asymmetry and phenotypic variance for the number of sternopleural bristles and in the length of two wing traits over three generations after hybridization. We found a 'hybrid vigour effect' in Fl females with an increase in fecundity relative to their parental populations. The F2 and F3 females showed increased fluctuating asymmetry in several traits and reduced fecundity compared with the Fl females, probably due to a breakdown in coadapted gene complexes. The males followed the same pattern of fluctuating asymmetry for bristles but there was no increase in wing fluctuating asymmetry in the F2 and F3 generations. Trait differences in phenotypic variance were found between wings and bristles. We found an increase in phenotypic variance in the Fl generation for both sexes and all traits, which could be due to increased genetic variance after hybridization. The phenotypic variance increased further in generations F2 and F3 for bristle number. For the wings, phenotypic variance generally decreased in generations F2 and F3 when compared with Fl, which we attribute to canalization and selection on the wings.

Variation in genetic and environmental influences in serum lipid and apolipoprotein levels across the lifespan in Swedish male and female twins

The contribution of genetic and environmental factors to variation in lipids and apolipoproteins has been estimated in previous twin and family studies. However, it is unclear whether there are sex and/or age differences in parameter estimates. We investigated a sample selected from the population-based Swedish Twin Registry of 725 like- and unlike-sex twin pairs, ages 17-85. Quantitative genetic methods were used to evaluate sex and age differences in genetic and environmental variation in lipid and apolipoprotein levels in three age groups, 17-49, 50-69, and 70-85. Heritabilities for lipids and apolipoproteins ranged from 35%-74%. Consistent sex differences were found in triglycerides. Females had higher heritabilities (56%) than males (35%) across the age groups. Total phenotypic variation increased across the age groups for cholesterol and apolipoprotein B due to an increase in unique environmental variance components. In contrast, in apolipoprotein A1 variance was highest in the middle age group and no differences were found in the phenotypic variance between age groups for triglycerides. We concluded that differences in phenotypic variation for cholesterol and apolipoprotein B were almost entirely due to the accumulation of environmental experiences throughout life, whereas there were no consistent patterns of differences in phenotypic variance for apolipoprotein A1 and triglycerides.

Variation in pelvic size between males and females in nonhuman anthropoids

Whether there is a sexual difference in phenotypic variance has been the subject of theoretical and empirical studies. The presence of such a difference is integral to some models on the evolution of sexual dimorphism. Several studies report that males are more variable than females for nonpelvic measures. This study tests for a sexual difference in variability of the pelvis. Phenotypic variance is a correlate of both the intensity and mode of selection. In both sexes, the pelvis is subject primarily to stabilizing selection. However, selection intensity is greater among females than among males because only among females does the pelvis function as a birth canal and, thereby, serve as a proximate cause of death. As selection intensity and phenotypic variability are inversely related under stabilizing selection, the implication is that females should be less variable than males in pelvic size and shape. However, the results of previous studies on this issue are equivocal or contradictory. This study compares the sexes for differences in pelvic variability by analyzing nine species (and ten samples) of nonhuman anthropoids. Data were collected on 16 measures of the pelvis; only adults were used. Levene's univariate and multivariate tests for relative variation were used in the analysis. The results show that the sexes do not differ significantly in pelvic variability either within or among the nine samples of noncaptive anthropoids. Only in the one sample of captive specimens (Saimiri sciureus) do the sexes differ significantly in pelvic variability, with males being more variable than females. Two interpretations are derived from these results. First, the sexual difference in variability in captive Saimiri may be due either to a bias in the selection of Saimiri specimens by the captors/experimentors or to a sexual difference in growth associated with the stress of captivity. Second, the sexes do not differ in pelvic variability among noncaptive anthropoids because the anlage of the pelvis is bipotential in development. The intrinsic (i.e., agonadal) pattern of growth and development is that characteristic of a female. Testicular androgens are requisite to redirect the pattern of growth and development to that characteristic of a male. However, the redirected pattern of growth in males simply shifts en masse the intrinsic (i.e., female) distribution curve. The consequence among adults is sexual dimorphism in pelvic size, but sexual equivalency in relative pelvic variability.

Effect of accounting for different phenotypic variances by sire breed and sex on selection of sires based on expected progeny differences for 200- and 365-day weights.

The effects of accounting for different phenotypic variances according to sire breed and sex subclasses on estimation of sire breed effects and prediction of expected progeny differences of sires mated to Hereford and Angus cows were investigated. Data consisted of 6,977 and 6,530 records of 200-d (weaning) and 365-d (yearling) weights, respectively, of F1 calves sired by bulls (662 and 661, respectively) of 23 breeds that have been evaluated in the Germ Plasm Evaluation Program at the U.S. Meat Animal Research Center, Clay Center, NE. Models compared included fixed effects of genetic group of sire (samples of sires evaluated at different times), dam breed, sex, birth year of calf and age of dam, plus sire within genetic group and dam within dam breed as random effects. Variance structures were different: Model I assumed homogeneous variances across sire breed-sex subclasses; Model II accounted for differences in phenotypic variance by sire breed and sex subclasses. Differences between estimates of sire group effects obtained with the two models were not significant for either trait. Product-moment and rank correlations between expected progeny differences obtained with Model I and Model II were greater than .93 when computed within each group and .99 or larger when computed across breeds. There were slight changes in the numbers of sires contributed by different breeds to the proportions selected across breeds under different selection intensities when sires were ranked with the two models. However, differences between means predicted under Model II were small when sires were ranked and selected based on the two models. Changes in standard errors of prediction for expected progeny differences and standard errors for estimates of breed effects obtained when adjusting for differences of phenotypic variances, compared to not adjusting, were proportional to the ratios of the phenotypic standard deviations of the sire breeds to the common phenotypic standard deviation.

Selection Among S1 Families Vs. Selfed Half‐Sib or Full‐Sib Families in Autogamous Crops

In autogamous craps, half‐ and full‐sib families are rautinely generated in various systems of intermating. Usually, these families have not been used as units of selection, due to low expected gains relative to other selection units (e.g., S1 families). The use of selfed half‐ (SHS) or full‐sib (SFS) families, and differences in phenotypic variance between these and S1 families, might reduce the difference in expected gains between the selection methods. Our primary objective was to predict the expected gain fram selection on each family unit, based on a genetic model with additive and homozygous dominance effects. A second objective was to empirically estimate phenotypic variances among SHS, SFS, and S1 families derived fram random‐mating populations of wheat (Triticum aestirum L.) and soybean [Glycine max (L.) Merr.], and use those estimates to compare expected gains. Estimates of phenotypic variance for yield of SFS wheat families (O2SFS) were small compared with estimates for S1 families (O2S1) due to smaller error variances. Phenotypic variances of SHS soybean families (O2SHS) tended to be smaller than those of S1 families for seed yield and size but not for seed pratein and oil concentrations. As with the wheat populations, error variances tended to be smaller for SHS families. The expected relative efficiencies of S1 family selection vs. SFS or SHS family selection are 1.14 (σSFS/σS1) in wheat and 2.29 (σSHS/σS1) soybean. Substituting phenotypic standard deviations into each formula resulted in no consistent advantage to using S1 family selection with the exception of seed pratein and oil in soybean. In conclusion, election among SHS or SFS families should be of greatest benefit for productivity traits that require larger seed quantifies for testing. These selection units allow larger plot sizes and/or increased replications, thereby impraving the precision of entry‐mcan estimates and reducing phenotypic variance.

Aspects of the population genetics of Mytilus (Mytilidae; Mollusca) in the British Isles

Starch gel electrophoresis was used to study variation at 11 loci in mussels sampled mainly from British coastal sites. Two types of mussel were identified, Mytilus edulis, the common mussel and its southern relative Mytilus galloprovincialis. Several partially diagnostic loci were used to map the distribution of the two forms. Mytilus edulis was present at all sites sampled in Britain and Ireland but was at low frequency in SW England; M. galloprovincialis was detected in SW England, the south and west of Ireland. Scotland and NE England, but was absent from south Wales, the Irish sea coasts of Wales and Ireland, and SE England. Apart from the occurrence of M. galloprovincialis in NE England, this distribution conforms with the results of studies using morphological characters and parallels the distribution of many other southern species in Britain. At the microgeographical level, M. edulis was found to prefer more sheltered and estuarine conditions than M. galloprovincialis. Analysis using the best diagnostic loci showed that hybridization is occurring between M. edulis and M. galloprovincialis at all localities where they occur sympatrically but that the extent of hybridization varies considerably between localities. The distribution of localities having high proportions of hybrid individuals is best interpreted by assuming that hybrids have higher fitness than parental types at these localities. A study was made of variation within and between those localities where only M. edulis individuals were observed. Little significant geographic variation in allele frequency was detected, but significant deficits of heterozygotes compared with Hardy-Weinberg expectations were seen for most loci. Analysis suggests that the Wahlund effect is not involved and that the most likely cause of the deficit is low frequencies of null alleles. In M. edulis no differences in phenotypic variance in shell height and width were observed between samples of multiply heterozygous and multiply homozygous individuals and no genetic differences were found between juveniles and adults. Overall little evidence was found that balancing selection is responsible for maintenance of the polymorphisms studied in M. edulis. The pattern of geographic variation in gene frequencies in Mytilus in the British Isles is discussed in relation to variation in the south and north of Europe and North America. It is concluded that steep clines in gene frequencies in M. edulis observed by other workers in the Baltic and in Long Island Sound cannot be attributed to the presence of M. galloprovincialis.