Sequence-based variations in starch synthesis-related genes (SSRGs) exert a basic influence on the determination of eating quality in rice (Oryza sativa L.). This study aimed to investigate the relationship between the sequence variations from parts of 10 SSRGs and the amylose content (AC) plus rapid viscosity analysis (RVA) profiles in a heuristic rice core set by association mapping (AM). In total, 86 sequence variations were found in 10 sequenced amplicons, including 79 single-nucleotide polymorphisms (SNPs), six insertion-deletions (InDels) and one polymorphic simple sequence repeat (SSR). Among them, 61 variations were exon-based, of which 41 should lead to amino acid changes. Four subpopulations were revealed by population structure analysis based on 170 genome-wide SSR genotypes. The final AM showed a sum of four significant associations between three phenotypic indices and three sequence variations. An ADP-glucose pyrophosphorylase small unit 1 (OsAGPS1) SNP (A–G) was significantly associated with increased AC (P<0·001, R2=15·6%), while a 12-bp deletion of AGPase large unit 4 (OsAGPL4) was significantly related to the decreased breakdown viscosity (BDV) (P<0·001, R2=16·6%) in both general linear model (GLM) and mixed linear model (MLM). This study provides a new perspective of allele mining for breeding strategies based on marker-assisted selection.

Dwarfing in rice has dramatically improved and stabilized rice yields worldwide, often controlled by a single dwarf gene, sd1. A novel semidwarf gene d60 complements the gametic lethal gene gal, such that the F1 between ‘Hokuriku 100’ (genotype d60d60GalGal, Gal: mutant non-lethal allele) and ‘Koshihikari’ (D60D60galgal, D60: tall allele) would show 25% sterility due to deterioration of gametes bearing both gal and d60. The F2 would segregate as one semidwarf (1 d60d60GalGal) : two tall and 25% sterile (2 D60d60Galgal) : six tall (2 D60d60GalGal : 1 D60D60GalGal : 2 D60D60Galgal : 1 D60D60galgal), skewed from a Mendelian segregation ratio of one semidwarf : three tall for a single recessive gene. To pyramid d60 and sd1, into the Japanese super-variety ‘Koshihikari’, the F1 (D60d60Galgal) of ‘Koshihikari’ × ‘Hokuriku 100’ was first backcrossed with ‘Koshihikari’, and the BCF1 segregated into a ratio of one tall and 25% sterile (D60d60Galgal) : two tall (1 D60D60Galgal : 1 D60D60galgal). Tall, 25% sterile BC1F1 plants (D60d60Galgal) were then selected for pollen sterility and backcrossed with ‘Koshihikari’ as the recurrent parent. It is unnecessary to grow out and select a semidwarf from the BCnF2 if a pollen parent with ∼70% pollen fertility is chosen from the BCnF1 to backcross with the recurrent parent. Semidwarfing genes d60 and sd1 were successfully pyramided into the ‘Koshihikari’ genome by crossing isogenic lines ‘Koshihikari d60’ and ‘Koshihikari sd1’, to produce ‘Minihikari’, a new parental source of both d60 and sd1. ‘Minihikari’ displayed super-short stature due to the combination of sd1 and d60, which are genetically and functionally independent.

Dry matter grain yield per plot from three genetically homogeneous single-cross maize hybrids were analysed to investigate whether environmental variance depends on genotype. Three genotypes were tested at 20 locations in 3 years. The data were analysed using a non-parametric approach and fully parametric Bayesian models. Both analyses reveal effects of genotype on environmental variation. The Bayesian analyses indicate that genotype by location–year interactions are the most important effects acting at the level of the mean. The best-fitting Bayesian model is one postulating genotype by location–year interactions acting on the mean and main effects of genotype and of location–year on the variance. Despite the detection of genotypic effects acting on the variance, location–year effects constitute the biggest relative source of variance heterogeneity.

Intrinsic postzygotic reproductive isolation is thought to result from the substitution of multiple harmless or beneficial genetic differences between species that are incidentally deleterious when combined in species hybrids, causing hybrid sterility or inviability. Genetic variability for hybrid sterility or inviability phenotypes is, however, rarely assessed in natural populations. Here, we assess variation for Drosophila simulans-encoded maternal factor(s) that cause lethality in D. simulans–Drosophila melanogaster F1 hybrid females. First, we survey genetic variability in the strength of D. simulans-mediated maternal effect hybrid lethality among 37 geographic and laboratory isolates. We find abundant variability in the strength of maternal effect hybrid lethality, ranging from complete lethality to none. Second, we assess maternal effect hybrid lethality for a subset of wild isolates made heterozygous with two so-called hybrid rescue strains. The results suggest that the D. simulans maternal effect hybrid lethality involves a diversity of alleles and/or multiple loci.

Heterosis is widely used in genetic crop improvement; however, the genetic basis of heterosis is incompletely understood. The use of whole-genome segregating populations poses a problem for establishing the genetic basis of heterosis, in that interactions often mask the effects of individual loci. However, introgression line (IL) populations permit the partitioning of heterosis into defined genomic regions, eliminating a major part of the genome-wide epistasis. In our previous study, based on mid-parental heterosis (HMP) value with single-point analysis, 42 heterotic loci (HLs) associated with six yield-related traits were detected in wild and cultivated rice using a set of 265 ILs of Dongxiang common wild rice (Oryza rufipogon Griff.). In this study, the genetic effects of HLs were determined as the combined effects of both additive and dominant gene actions, estimated from the performance values of testcross F1s and the dominance effects estimated from the HMP values of testcross F1s. We characterized the gene action type at each HL. Thirty-eight of the 42 HLs were over-dominant, and in the absence of epistasis, four HLs were dominant. Therefore, we favour that over-dominance is a major genetic basis of ‘wild-cultivar’ crosses at the single functional Mendelian locus level.

Mapping of quantitative trait loci (QTLs) underlying variation in quantitative traits continues to be a powerful tool in genetic study of plants and other organisms. Whole genome average interval mapping (WGAIM), a mixed model QTL mapping approach using all intervals or markers simultaneously, has been demonstrated to outperform composite interval mapping, a common approach for QTL analysis. However, the advent of high-throughput high-dimensional marker platforms provides a challenge. To overcome this, a dimension reduction technique is proposed for WGAIM for efficient analysis of a large number of markers. This approach results in reduced computing time as it is dependent on the number of genetic lines (or individuals) rather than the number of intervals (or markers). The approach allows for the full set of potential QTL effects to be recovered. A proposed random effects version of WGAIM aims to reduce bias in the estimated size of QTL effects. Lastly, the two-stage outlier procedure used in WGAIM is replaced by a single stage approach to reduce possible bias in the selection of putative QTL in both WGAIM and the random effects version. Simulation is used to demonstrate the efficiency of the dimension reduction approach as well as demonstrate that while the approaches are very similar, the random WGAIM performs better than the original and modified fixed WGAIM by reducing bias and in terms of mean square error of prediction of estimated QTL effects. Finally, an analysis of a doubled haploid population is used to illustrate the three approaches.

Two sets of reciprocal introgression lines (ILs) and a population of recombinant inbred lines (RILs) derived from the cross between japonica cultivar Xiushui09 and indica breeding line IR2061-520-6-9 (abbreviated as IR2061) were used to identify QTL for heading date (HD). Phenotyping was conducted in Hainan Island for two winter seasons (2007 and 2009). Nine QTLs were detected in the ILs with Xiushui09 background (XS-ILs), and four of which were repeatedly mapped across 2 years. Five QTLs were identified in the ILs with IR2061 background (IR-ILs), and three of which were commonly detected in 2 years. All commonly detected QTL had the same direction of gene effect. Seven QTL for HD were identified in the RILs in 2009. Only three (25%) QTLs were commonly detected using all the three populations (XS-ILs, IR-ILs and RILs). The number of commonly identified QTLs among populations was related to degree of similarity of their genetic backgrounds, suggesting that the genetic background effect is important for detecting HD QTL. QHd7 and QHd10b stably expressed in different populations and across years thus would be exploited in rice breeding programme. Moreover, lines with both of QHd7 and QHd10b resulted in at least 3 days earlier than lines with only one of them QTL, showing evident pyramiding effect.

Genome wide association studies (GWAS) have largely succeeded family-based linkage studies in livestock and human populations as the preferred method to map loci for complex or quantitative traits. However, the type of results produced by the two analyses contrast sharply due to differences in linkage disequilibrium (LD) imposed by the design of studies. In this paper, we demonstrate that association and linkage studies are in agreement provided that (i) the effects from both studies are estimated appropriately as random effects, (ii) all markers are fitted simultaneously and (iii) appropriate adjustments are made for the differences in LD between the study designs. We demonstrate with real data that linkage results can be predicted by the sum of association effects. Our association study captured most of the linkage information because we could predict the linkage results with moderate accuracy. We suggest that the ability of common single nucleotide polymorphism (SNP) to capture the genetic variance in a population will depend on the effective population size of the study organism. The results provide further evidence for many loci of small effect underlying complex traits. The analysis suggests a more informed method for GWAS is to fit statistical models where all SNPs are analysed simultaneously and as random effects.

The possibility of breeding for uniform individuals by selecting animals expressing a small response to environment has been studied extensively in animal breeding. Bayesian methods for fitting models with genetic components in the residual variance have been developed for this purpose, but have limitations due to the computational demands. We use the hierarchical (h)-likelihood from the theory of double hierarchical generalized linear models (DHGLM) to derive an estimation algorithm that is computationally feasible for large datasets. Random effects for both the mean and residual variance parts of the model are estimated together with their variance/covariance components. An important feature of the algorithm is that it can fit a correlation between the random effects for mean and variance. An h-likelihood estimator is implemented in the R software and an iterative reweighted least square (IRWLS) approximation of the h-likelihood is implemented using ASReml. The difference in variance component estimates between the two implementations is investigated, as well as the potential bias of the methods, using simulations. IRWLS gives the same results as h-likelihood in simple cases with no severe indication of bias. For more complex cases, only IRWLS could be used, and bias did appear. The IRWLS is applied on the pig litter size data previously analysed by Sorensen & Waagepetersen (2003) using Bayesian methodology. The estimates we obtained by using IRWLS are similar to theirs, with the estimated correlation between the random genetic effects being −0·52 for IRWLS and −0·62 in Sorensen & Waagepetersen (2003).

Fertility transmission (FT) is a phenomenon with a cultural and/or genetic basis, whereby a positive correlation exists between the number of offspring of an individual and that of his/her parents. Theoretical studies using a haploid individual-based model have shown that FT increases the variance and intergenerational correlation in reproductive success and results in an imbalance in the coalescent tree of sampled genes. This phenomenon has been documented in several demographic studies conducted on the correlation in fertility between generations, or through the reconstruction of the genealogical trees of mitochondrial DNA sequences. However, as mtDNA is a single locus, potentially subject to other forces (e.g. natural selection), it is of interest to extend the theory of FT to nuclear loci. We show that because random mating between individuals leads to a mixing of their fertility profiles, FT in these cases will have less influence on the variance and intergenerational correlation of reproductive success. This, in turn, results in less impact on the shape of the coalescent trees. Nevertheless, in the presence of FT, high heterogeneity in reproductive success and homogamy for family size will increase the imbalance in the coalescent tree. Thus, FT should be easier to detect when occurring in conjunction with these other factors. We also show the utility of analysing different kinds of loci (X-linked, Y-linked, mitochondrial and autosomal) to assess whether FT is matrilineal, patrilineal or biparental. Finally, we demonstrate that the shape of the coalescent tree depends upon population size, in contrast to the classical Kingman's model.

Seed filling rate of soybean has been shown to be a dynamic process in different developmental stages affected by both genotype and environment. The objective of the present study was to determine additive, epistatic and quantitative trait loci (QTLs)×environment interaction (QE) effects of the QTL underlying a seed filling rate of soybean. One hundred and forty-three recombinant inbred lines (RILs) derived from the cross of Charleston and Dongnong 594 were used with 2 years of field data (2004 and 2005). Eleven QTLs with significantly unconditional and conditional additive (a) effect and/or additive×environment interaction (ae) effect at different filling stages were identified. Of them six QTLs showed positive a effects and four QTLs had negative a effects on the seed filling rate during seed development. aa and aae effects of 12 pairs of QTLs were identified by unconditional mapping from the initial stage to the final stage. Thirteen pairs of QTLs underlying the seed filling rate with aa and aae effects were identified by conditional mapping. QTLs with aa and aae (additive×additive×environment) effects appeared to vary at different filling stages. Our results demonstrated that the mass filling rate in soybean seed were under genetic and environmental control.

Epistasis is an important feature of the genetic architecture of quantitative traits, but the dynamics of epistatic interactions in natural populations and the relationship between epistasis and pleiotropy remain poorly understood. Here, we studied the effects of epistatic modifiers that segregate in a wild-derived Drosophila melanogaster population on the mutational effects of P-element insertions in Semaphorin-5C (Sema-5c) and Calreticulin (Crc), pleiotropic genes that affect olfactory behaviour and startle behaviour and, in the case of Crc, sleep phenotypes. We introduced Canton-S B (CSB) third chromosomes with or without a P-element insertion at the Crc or Sema-5c locus in multiple wild-derived inbred lines of the Drosophila melanogaster Genetic Reference Panel (DGRP) and assessed the effects of epistasis on the olfactory response to benzaldehyde and, for Crc, also on sleep. In each case, we found substantial epistasis and significant variation in the magnitude of epistasis. The predominant direction of epistatic effects was to suppress the mutant phenotype. These observations support a previous study on startle behaviour using the same D. melanogaster chromosome substitution lines, which concluded that suppressing epistasis may buffer the effects of new mutations. However, epistatic effects are not correlated among the different phenotypes. Thus, suppressing epistasis appears to be a pervasive general feature of natural populations to protect against the effects of new mutations, but different epistatic interactions modulate different phenotypes affected by mutations at the same pleiotropic gene.

A common problem for genome-wide association analysis (GWAS) is lack of power for detection of quantitative trait loci (QTLs) and precision for fine mapping. Here, we present a statistical method, termed single-step GBLUP (ssGBLUP), which increases both power and precision without increasing genotyping costs by taking advantage of phenotypes from other related and unrelated subjects. The procedure achieves these goals by blending traditional pedigree relationships with those derived from genetic markers, and by conversion of estimated breeding values (EBVs) to marker effects and weights. Additionally, the application of mixed model approaches allow for both simple and complex analyses that involve multiple traits and confounding factors, such as environmental, epigenetic or maternal environmental effects. Efficiency of the method was examined using simulations with 15 800 subjects, of which 1500 were genotyped. Thirty QTLs were simulated across genome and assumed heritability was 0·5. Comparisons included ssGBLUP applied directly to phenotypes, BayesB and classical GWAS (CGWAS) with deregressed proofs. An average accuracy of prediction 0·89 was obtained by ssGBLUP after one iteration, which was 0·01 higher than by BayesB. Power and precision for GWAS applications were evaluated by the correlation between true QTL effects and the sum of m adjacent single nucleotide polymorphism (SNP) effects. The highest correlations were 0·82 and 0·74 for ssGBLUP and CGWAS with m=8, and 0·83 for BayesB with m=16. Standard deviations of the correlations across replicates were several times higher in BayesB than in ssGBLUP. The ssGBLUP method with marker weights is faster, more accurate and easier to implement for GWAS applications without computing pseudo-data.

Current methods for detecting Hardy–Weinberg disequilibrium (HWD) only deal with one locus at a time. We developed a method that can jointly detect HWD for multiple loci. The method was developed under the generalized linear model (GLM) using the probit link function. When applied to a single locus, the new method is more powerful than the exact test. When applied to two or more loci, the method can reduce false positives caused by linkage disequilibrium (LD). We applied the method to 24 single nucleotide polymorphism (SNP) markers of a single human gene and eliminated several false positive HWDs due to LD. We developed an R package ‘hwdglm’ for joint HWD detection, which can be downloaded from our personal website (www.statgen.ucr.edu).

Imputation of moderate-density genotypes from low-density panels is of increasing interest in genomic selection, because it can dramatically reduce genotyping costs. Several imputation software packages have been developed, but they vary in imputation accuracy, and imputed genotypes may be inconsistent among methods. An AdaBoost-like approach is proposed to combine imputation results from several independent software packages, i.e. Beagle(v3.3), IMPUTE(v2.0), fastPHASE(v1.4), AlphaImpute, findhap(v2) and Fimpute(v2), with each package serving as a basic classifier in an ensemble-based system. The ensemble-based method computes weights sequentially for all classifiers, and combines results from component methods via weighted majority ‘voting’ to determine unknown genotypes. The data included 3078 registered Angus cattle, each genotyped with the Illumina BovineSNP50 BeadChip. SNP genotypes on three chromosomes (BTA1, BTA16 and BTA28) were used to compare imputation accuracy among methods, and the application involved the imputation of 50K genotypes covering 29 chromosomes based on a set of 5K genotypes. Beagle and Fimpute had the greatest accuracy among the six imputation packages, which ranged from 0·8677 to 0·9858. The proposed ensemble method was better than any of these packages, but the sequence of independent classifiers in the voting scheme affected imputation accuracy. The ensemble systems yielding the best imputation accuracies were those that had Beagle as first classifier, followed by one or two methods that utilized pedigree information. A salient feature of the proposed ensemble method is that it can solve imputation inconsistencies among different imputation methods, hence leading to a more reliable system for imputing genotypes relative to independent methods.

Genomic selection refers to the use of dense, genome-wide markers for the prediction of breeding values (BV) and subsequent selection of breeding individuals. It has become a standard tool in livestock and plant breeding for accelerating genetic gain. The core of genomic selection is the prediction of a large number of marker effects from a limited number of observations. Various Bayesian methods that successfully cope with this challenge are known. Until now, the main research emphasis has been on additive genetic effects. Dominance coefficients of quantitative trait loci (QTLs), however, can also be large, even if dominance variance and inbreeding depression are relatively small. Considering dominance might contribute to the accuracy of genomic selection and serve as a guide for choosing mating pairs with good combining abilities. A general hierarchical Bayesian model for genomic selection that can realistically account for dominance is introduced. Several submodels are proposed and compared with respect to their ability to predict genomic BV, dominance deviations and genotypic values (GV) by stochastic simulation. These submodels differ in the way the dependency between additive and dominance effects is modelled. Depending on the marker panel, the inclusion of dominance effects increased the accuracy of GV by about 17% and the accuracy of genomic BV by 2% in the offspring. Furthermore, it slowed down the decrease of the accuracies in subsequent generations. It was possible to obtain accurate estimates of GV, which enables mate selection programmes.

High-throughput gene profiling studies have been extensively conducted, searching for markers associated with cancer development and progression. In this study, we analyse cancer prognosis studies with right censored survival responses. With gene expression data, we adopt the weighted gene co-expression network analysis (WGCNA) to describe the interplay among genes. In network analysis, nodes represent genes. There are subsets of nodes, called modules, which are tightly connected to each other. Genes within the same modules tend to have co-regulated biological functions. For cancer prognosis data with gene expression measurements, our goal is to identify cancer markers, while properly accounting for the network module structure. A two-step sparse boosting approach, called Network Sparse Boosting (NSBoost), is proposed for marker selection. In the first step, for each module separately, we use a sparse boosting approach for within-module marker selection and construct module-level ‘super markers’. In the second step, we use the super markers to represent the effects of all genes within the same modules and conduct module-level selection using a sparse boosting approach. Simulation study shows that NSBoost can more accurately identify cancer-associated genes and modules than alternatives. In the analysis of breast cancer and lymphoma prognosis studies, NSBoost identifies genes with important biological implications. It outperforms alternatives including the boosting and penalization approaches by identifying a smaller number of genes/modules and/or having better prediction performance.

A major challenge in complex trait genetics is to unravel how multiple loci and environmental factors together cause phenotypic diversity. Both first (F1) and second (F2) generation hybrids often display phenotypes that deviate from what is expected under intermediate inheritance. We have here studied two chicken F2 populations generated by crossing divergent chicken lines to assess how epistatic loci, identified in earlier quantitative trait locus (QTL) studies, contribute to hybrid deviations from the mid-parent phenotype. Empirical evidence suggests that the average phenotypes of the intercross birds tend to be lower than the midpoint between the parental means in both crosses. Our results confirm that epistatic interactions, despite a relatively small contribution to the phenotypic variance, play an important role in the deviation of hybrid phenotypes from the mid-parent values (i.e. multi-locus hybrid genotypes lead to lower rather than higher body weights). To a lesser extent, dominance also appears to contribute to the mid-parent deviation, at least in one of the crosses. This observation coincides with the hypothesis that hybridization tends to break up co-adapted gene complexes, i.e. generate Bateson–Dobzhansky–Muller incompatibilities.

In this paper, we developed and compared several expectation–maximization (EM) algorithms to find maximum likelihood estimates of individual inbreeding coefficients using molecular marker information. The first method estimates the inbreeding coefficient for a single individual and assumes that allele frequencies are known without error. The second method jointly estimates inbreeding coefficients and allele frequencies for a set of individuals that have been genotyped at several loci. The third method generalizes the second method to include the case in which null alleles may be present. In particular, it is able to jointly estimate individual inbreeding coefficients and allele frequencies, including the frequencies of null alleles, and accounts for missing data. We compared our methods with several other estimation procedures using simulated data and found that our methods perform well. The maximum likelihood estimators consistently gave among the lowest root-mean-square-error (RMSE) of all the estimators that were compared. Our estimator that accounts for null alleles performed particularly well and was able to tease apart the effects of null alleles, randomly missing genotypes and differing degrees of inbreeding among members of the datasets we analysed. To illustrate the performance of our estimators, we analysed previously published datasets on mice (Mus musculus) and white-tailed deer (Odocoileus virginianus).

In previous analyses, the variation in actual, or realized, relationship has been derived as a function of map length of chromosomes and type of relationship, the variation being greater the shorter the total chromosome length and the coefficient of variation being greater the more distant the relationship. Here, the results are extended to allow for the relatives’ ancestor being inbred. Inbreeding of a parent reduces variation in actual relationship among its offspring, by an amount that depends on the inbreeding level and the type of mating that led to that level. For descendants of full-sibs, the variation is reduced in later generations, but for descendants of half-sibs, it is increased.