Meta-Analysis of Influence of Diversity of Parental Forms on Heterosis and Specific Combining Ability of Their Hybrids

Featured Application

Information on the genetic potential of parental forms can shorten the breeding process. Knowledge of the variation of parental forms at the phenotypic and genetic levels can have a very significant impact on the heterosis effect of hybrids.

Abstract

An important stage in any breeding activity is selection of suitable individuals for further breeding. Thus, the main goal of breeders becomes such a selection of parental forms that leads to the consolidation and maximization of the value of traits of significant utility and economic importance. Heterosis and specific combining ability are very important parameters in plant and animal breeding. The ability to predict their value and relevance could significantly shorten the breeding process. One way to predict the effects of heterosis and specific combining ability is to select parental forms for crosses. This selection can be made on the basis of variation in parental forms. An analysis was made of publicly available data that contain information about the effects of heterosis, the effects of specific combining ability, and phenotypic and genetic diversity of parental forms. Preliminary studies show that the best approach for obtaining favorable hybrids would be selection of parental forms that are very genetically diverse while being phenotypically equal.

1. Introduction

Selection of suitable individuals for further breeding is one of the stages of any breeding work. Selection of parental forms leading to perpetuation and maximization of the value of traits of significant utility and economic importance is the main goal of breeders. Parameters of great importance in plant and animal breeding are heterosis and specific combining ability (SCA) [1,2,3]. Heterosis is the increased exuberance, vigor, and fertility of hybrids compared to the initial parental forms, as a result of the crosses from which they were derived [4]. Underlying heterosis are phenomena of gene interaction that, for example, through the occurrence of heterozygous systems and, in particular, through dominant systems result in better adaptation of hybrids to the environment or outright greater productivity [5,6,7]. Heterosis plays a key role in plant breeding and is important in both agriculture and research. In plant breeding, heterosis is used to produce varieties with higher yields, better quality, and resistance to diseases and pests. With heterosis, it is also possible to improve the adaptability of plants to different environmental conditions, such as drought, soil salinity, and temperature extremes. Hybrid varieties often show greater disease resistance because they can have different alleles of resistance genes from both parents. For many crops, such as corn, wheat, and rice, the use of hybrid varieties is common and contributes to higher yields. In research, heterosis is used to study trait inheritance and population genetics and to understand plant development and evolution processes. Hybrid varieties can also help increase genetic diversity in plant populations, which is important for long-term crop stability and resistance to changing environmental conditions. The use of heterosis can help reduce the cost of crop production, as hybrid varieties can show greater efficiency in the use of resources, such as water and fertilizer. Heterosis is particularly important in breeding annual crops, which are often grown on a large scale to produce food for a growing world population. The use of heterosis in plant breeding can lead to a reduction in the use of pesticides and herbicides, as hybrid varieties often show greater resistance to pests and weeds. Hybrid varieties can also have a positive impact on the development of sustainable agriculture, by increasing production efficiency and reducing negative environmental impacts. Conclusions from research on heterosis in plant breeding can be applied to improve breeding strategies and increase the efficiency of food production to meet growing social needs.

Specific combining ability (SCA) is a type of combining ability that is based on non-additive allelic effects [8]. It is determined by the difference between the trait value in a given cross relative to the expected value determined on the basis of overall combining ability [9]. Specific combining ability plays a key role in plant breeding and is important for obtaining desired traits and improving yields. SCA refers to the ability of certain plants to form efficient and desirable genetic combinations when crossbreeding. In plant breeding, SCA is used to produce hybrids with higher yield, quality, and resistance to diseases and pests. Plants with high SCA can be efficiently crossed with different varieties, resulting in progeny with desirable traits. The use of SCA allows selection and combining of resistance genes, which contributes to plants that are less susceptible to diseases and pests. SCA is particularly important for annual crops, such as corn and wheat, which are grown on a large scale for food production. The use of SCA allows plants with different maturity dates to be crossed, allowing yields to be adjusted for different seasons. SCA helps increase genetic diversity in plant populations, which is important for long-term crop stability and resistance to changing environmental conditions. Growers are using SCA to produce plants with higher fertility, which translates into higher profits for producers. The use of SCAs can help reduce crop production costs by increasing efficiency in the use of resources such as water and fertilizer. For autogamous crops, specific combining ability can be important to increase genetic variability and avoid endogamous depression. SCA is also important in genetic research, allowing analysis of genotype interactions and identification of genes responsible for desirable traits. SCA can be used to produce plants with more stable growth and development, which affects yield uniformity.

The ability to predict the value and relevance of heterosis and SCA effects could shorten the breeding process; in some cases even by several years. This would translate into obvious economic benefits, among others. One possibility for such a potential prediction of heterosis and SCA effects is selection of parental forms for crosses. We can perform this selection based on the diversity of parental forms: phenotypic or genetic [10,11,12,13,14].

Phenotypic variation depends on: the measure of its estimation used, environmental conditions, selection/choice of observed traits, and the number of traits analyzed. Environmental conditions appear to be the factor that most significantly determines the responses of genotypes to observed traits. Golinski et al. [15], studying winter wheat cultivars infected with Fusarium graminearum, observed very different responses of two pairs of cultivars, Success and Rywalka, as well as Finezja and Soraja, in two localities: Cerekwica and Sitaniec.

Genotype is independent of environmental conditions, hence when assessing genetic diversity, this problem is eliminated. Genetic diversity, on the other hand, depends on the number of markers from which it is estimated. Recently, with the very large number of markers available, this problem seems less significant [16,17]. In addition, values of genetic similarity/diversity depend on the measure used to estimate it. Relationships between phenotypic and genetic diversity of eighteen parental lines of F1 hybrids of winter oilseed rape based on the CMS ogura hybridization system were examined by Kozak et al. [18]. The conclusion of this study was that genetic divergence is not the same as phenotypic divergence [18].

Considering the above, it may be interesting to answer the question: does the diversity of parental forms affect the values of the heterosis effect and the SCA effect? If there is such an influence, then another question can be posed: what kind of influence is it? To answer these two questions, publicly available data were analyzed in the context of the influence of diversity of parental forms on the heterosis effect and specific combining ability. Accordingly, the research hypothesis was that phenotypic and genetic variations in parental forms affect the values of heterosis and specific combining ability effects of hybrids obtained from crossing these parental forms. Studies to date tend to focus on analyzing a single population and evaluating breeding parameters such as general combining ability, SCA, and heterosis [19]. Less often, phenotypic or genetic variation of parental forms is analyzed in the context of evaluating hybrids obtained by crossing these parental forms [20,21,22,23,24]. Analyses of the determination of the differentiation of parental forms in terms of the values of breeding parameters of hybrids are not usually undertaken. In order to fill the gap in the literature on the subject, we investigated whether such an influence exists and, if so, what direction it takes.

The aim of this study was to analyze the effect of the diversity of parental forms on the values of breeding parameters, i.e., the heterosis effect and the SCA effect. The variation of parental forms was considered at two levels: phenotypic (based on Mahalanobis distance) and genetic (based on observations of molecular markers).

2. Materials and Methods

After analyzing 176 datasets from 38 articles [25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63], data were collected on:

(1)

heterosis effects:

$${\widehat{H}}_{ij}=\frac{{\overline{P}}_{ij}-{\overline{P}}_P}{{\overline{P}}_P},$$

(1)

where $H_{ij}$ denotes the heterosis effect of the ij-th hybrid, ${\overline{P}}_{ij}$ denotes the mean value of the ij-th hybrid, ${\overline{P}}_P$ denotes the mean value of the i-th and j-th parental forms.

(2)

SCA effects:

$${\widehat{SCA}}_{ij}=\frac{1}{2}\left(y_{ij}-y_{ji}\right)-{\widehat{g}}_i-{\widehat{g}}_j-\widehat{\mu },$$

(2)

where $SC{A}_{ij}$ denotes specific combining ability effect of the ij-th hybrid, $y_{ij}$ denotes the mean value for the genotype obtained by crossing the i-th and j-th parental forms, $g_i$ $\left(g_j\right)$ denotes the effect of general combining ability of the i-th (j-th) parental form, $\mu$ denotes the grand mean.

(3)

Phenotypic differentiation of parental forms was assessed by Mahalanobis distance:

$$d_{ij}={\left[\left({\mathit{y}}_i-{\mathit{y}}_j\right)\prime {\mathit{S}}^{-1}\left({\mathit{y}}_i-{\mathit{y}}_j\right)\right]}^{\raisebox{1ex}{$1$}\!\left/ \!\raisebox{-1ex}{$2$}\right.},$$

(3)

where $d_{ij}$ denotes Mahalanobis distance between the i-th and j-th hybrids, ${\mathit{y}}_i$ and ${\mathit{y}}_j$ denote the t-dimensional vectors of means of individual traits for the i-th and j-th parental forms, respectively, and S denotes the covariance matrix.

(4)

Genetic differentiation or similarity of parental forms:

$$G{S}_{ij}=\frac{2N_{ij}}{N_i+N_j},$$

(4)

where $N_i$ denotes the number of alleles in the i-th parental form, $N_j$ denotes the number of alleles in the j-th parental form, $N_{ij}$ denotes the number of alleles present in both the i-th and j-th parental forms.

The data collected included five crop species: maize [25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45], rapeseed [46,47,48,49,50,51,52], pea [53,54], barley [55,56,57], and wheat [58,59,60,61,62]. Datasets that met the following conditions were selected for analysis: (i) the effect of heterosis was calculated only in comparison with the average of the parental forms [63], (ii) Mahalanobis distances [64] were calculated for at least nine traits (t > 8), (iii) the diversity or genetic similarity of the parental forms was calculated according to the formula proposed by Nei and Li [65]. When the data were on genetic similarity (GS), it was converted to genetic diversity (GD) according to the formula:

GD = 1 − GS.

(5)

In total, 1980 sets of observations were analyzed. Relationships of heterosis and SCA effects from variation in parental forms were analyzed using Pearson’s linear correlation coefficients:

$$r_{XY}=\frac{cov\left(X,Y\right)}{{\sigma }_X{\sigma }_Y},$$

(6)

where $cov\left(X,Y\right)$ denotes the covariance between X and Y, ${\sigma }_X$ and ${\sigma }_Y$ denote the standard deviations of X and Y, respectively. Significance of coefficients of correlations (r) was tested using a two-sample t-test:

$$t=\frac{r\sqrt{n-2}}{\sqrt{1-r^2}}.$$

(7)

All these analyses were conducted using the GenStat v. 22 statistical software package [66].

3. Results

In the situations considered, the number of traits on the basis of which Mahalanobis distances of parental forms were calculated ranged from nine to nineteen. The number of markers on the basis of which genetic similarity/diversity was calculated ranged from 145 to 1703. Analyzing all 1980 sets of observations, statistically significant correlations were observed between genetic diversity and heterosis effects (−0.071) and Mahalanobis distances and SCA effects (0.049) (

Table 1. Pearson linear correlation coefficients and p-values (in brackets) between phenotypic distances (Mahalanobis) and genetic distances and heterosis and SCA effects calculated from all 1980 datasets.

Parameter	Phenotypic Distance	Genetic Distance
Heterosis effect	0.041 (0.074)	−0.071 (0.002)
SCA effect	0.049 (0.029)	0.012 (0.584)

).

The correlation coefficients had small absolute values despite statistical significance (Figure 1).

In addition, an intriguing result is the negative (−0.071) dependence of the heterosis effect on genetic diversity. The relationship between diversity of parental forms and breeding parameters may have been influenced by the fact that all cumulative effects—positive and negative, significant and insignificant—were included.

Let us consider only statistically significant heterosis effects. As before, and in this case, there was no significant effect of phenotypic variation and genetic variation of the parental forms on the heterosis effect values of their hybrids (Figure 2,

Table 2. Pearson’s linear correlation coefficients and p-values between phenotypic distances (Mahalanobis) and genetic distances and heterosis effects, for cases where heterosis effects were statistically significant.

Heterosis	Phenotypic Distance	Genetic Distance
Correlation coefficient	0.048	−0.013
p-value	0.518	0.886

).

An explanation for the above lack of correlation may be due to considering the cases of positive and negative heterosis effects together. Separating the two situations, a statistically significant effect of genetic diversity on heterosis effects was observed for both positive (r = 0.389, p < 0.001) and negative (r = −0.673, p < 0.001) heterosis effects (Figure 3 and Figure 4).

Phenotypic variation in parental forms did not affect the positive effects of heterosis (Figure 3), while it statistically significantly (r = 0.517, p < 0.001) affected the values of negative heterosis effects (Figure 4).

When considering the effect of diversity of parental forms on significant SCA effects, no statistical significance was observed at the phenotypic (r = 0.021, p = 0.606) and genetic (r = 0.027, p = 0.573) levels (Figure 5).

The results obtained (Figure 3 and Figure 4) indicate the influence of genetic diversity of parental forms on heterosis effects. Potential relationships were tested for the subset of data for which the genetic distances of parental forms were greater than 0.6 (Figure 6).

The results obtained do not indicate any statistical significance. This is probably due to the fact that all effects of heterosis were considered, without separating by their sign and significance.

The final stage of the study was to analyze the correlation between the values of genetic diversity of parental forms and the values of breeding parameters in situations when the effects of heterosis were statistically significant and when the genetic distances between parental forms were greater than 0.6. The results show that the genetic diversity of parental forms had a statistically significant effect on the positive (r = 0.326) and negative (r = −0.426) values of heterosis. In contrast, Mahalanobis distances between parental forms correlated (r = 0.305) with negative heterosis effects of their hybrids (

Table 3. Pearson linear correlation coefficients and p-values (in brackets) between phenotypic distances (Mahalanobis) and genetic distances and heterosis effects calculated when genetic distances were greater than 0.6 and heterosis effects were statistically significant.

Significant Effect of Heterosis	The Number of Significant Effects	Phenotypic Distance	Genetic Distance
Positive	68	0.190 (0.120)	0.326 (0.007)
Negative	46	0.305 (0.039)	−0.426 (0.003)

).

4. Discussion

The goal of every breeder is to obtain new varieties with the most favorable traits. For many years, attempts have been made through crossbreeding to obtain genotypes that would have higher yields and greater disease resistance than their parental forms. However, this is a very time-consuming and, unfortunately, inefficient process [67,68,69,70,71].

Selection and choosing individuals for mating is an important stage of breeding work. Some of the oldest and most common criteria determining selection of forms for further mating are phenotypic traits. Using the effects of general and specific combinatorial ability can lead to improved accuracy in predicting grain yield performance of hybrids. This accuracy can be improved by using the best linear unbiased prediction methods which are a promising tool for hybrid breeding programs [72]. Testing based on phenotypic traits can be subject to error due to the dependence of the expression of these traits on environmental conditions [73]. Diversity and population structure can vary significantly between morphological traits and molecular markers. Elucidating such discrepancies is crucial for further progress in breeding, as well as for maintaining genetic resources [49,50,51]. In order to achieve better breeding results, classical selection is increasingly being supplemented with test results using molecular markers linked to performance traits. Molecular markers based on DNA analysis are independent of environmental factors and show a high degree of polymorphism. The development of molecular techniques has revolutionized many areas of life and science. High-throughput genotyping and sequencing technologies make it possible to create predictive models that estimate the genetic value of the entire genotyped population. In a pre-breeding program, predictions speed up reproducible selection using rapid cycles in greenhouses by skipping some of the phenotyping steps [74]. The development of molecular techniques and genomics has allowed for a more complete understanding of the genetic basis of functional traits and thus the introduction of more informative techniques. The use of molecular markers in crop breeding is now a very convenient diagnostic tool in selection of favorable genotypes (MAS) [75]. Molecular markers assist breeding work during selection of an introduced gene or in selection of a genetic background. The most important advantages of MAS include the fact that selection of breeding material is primarily independent of the developmental stage of the plants under study and external factors [76,77]. Molecular markers appear to be a promising tool in predicting heterosis in various species, for example, in canola [78], maize [79,80], oats [81], rice [82], and wheat [14,83].

Physical mapping and association mapping make it possible to identify/select markers linked to genes determining quantitative traits, including the most important for breeders—yield [84,85]. The use of molecular markers in research has provided more accurate information about the non-allelic gene interaction—epistasis—which is so unfavorable during the selection process [6,86,87,88,89,90]. More recently, the interactions of higher orders of genes determining quantitative traits have begun to be explored [91,92,93], which can undoubtedly be of importance during selection processes conducted during breeding.

Association mapping is also conducted for hybrids of many plant species [94,95,96,97]. Recently, an attempt has been made to study the relationship between molecular markers and heterosis effects in maize [33]. The results show that the integration of molecular genetics with traditional methods of plant selection guarantees an objective assessment of the phenotype taking into account environmental factors and provides the opportunity to eliminate undesirable genotypes at the seedling stage, thus reducing breeding time.

Meta-analysis is a powerful statistical tool used in plant breeding that allows for the synthesis of results from multiple independent studies to obtain more precise and reliable conclusions. By analyzing data from different sources, meta-analysis enables plant breeders to understand general trends and relationships among plant traits. Meta-analysis can be used to compare different plant varieties in terms of yield, disease resistance, quality traits, and other breeding characteristics [98,99]. By synthesizing data from multiple studies, breeders can assess which variety has the highest breeding potential. Meta-analysis can help identify factors such as environmental conditions or cultivation methods that significantly impact plant performance [100,101,102]. Analyzing such data can lead to the optimization of breeding techniques. Meta-analysis can provide information on the effectiveness of different breeding strategies, such as selection, crossing, or genetic engineering [103,104]. Breeders can use these findings to develop optimal breeding strategies that lead to desired traits in plants. Meta-analysis can aid in assessing the stability of breeding traits under different environmental conditions [105,106]. Breeders can learn whether certain traits are more susceptible to environmental variability and which traits are consistently expressed. Meta-analysis can be used to evaluate the interaction between genotype and environment and their impact on plant performance [107,108,109]. Breeders can understand which varieties are the most stable in terms of performance under different environmental conditions. The insights gained from meta-analysis in plant breeding are valuable for developing better breeding strategies, increasing plant productivity, and developing varieties with desired traits. It is a powerful decision-support tool for plant breeders, allowing them to leverage existing scientific knowledge to develop more effective and sustainable breeding practices.

This paper showed the meta-analysis of the influence of diversity of parental forms on heterosis effect and specific combining ability. The results show the versatility of the marker data as a tool for diagnosing heterosis and specific combining ability effects. The advantage of the proposed approach is that it is independent of the heterosis effect marker. The positive effect of phenotypic diversity of parental forms on the negative effects of heterosis should be emphasized. With the concomitant negative effect of genetic diversity of parental forms, this indicates that the best way to obtain favorable hybrids would be to select parental forms that are very genetically diverse while being phenotypically equal. Relationships between genetic and phenotypic diversity of parental genotypes in barley were studied by Kuczyńska et al. [55]. They obtained results similar to those presented in our study. Similarly, Fufa et al. [61], studying hard red winter wheat genotypes, indicated the importance of conducting studies comparing phenotypic and genetic diversity. The determination of parameters related to combinatorial ability and heterosis from genetic variation in maize genotypes under stress and non-stress conditions is indicated in the study by Makumbi et al. [34]. Ali et al. [110] estimated the genetic distance between canola varieties using multivariate analysis. They investigated the relationship between genetic distance and heterosis. Their results indicated that the correlation between genetic distance and heterosis was positive and highly significant for seed yield, number of plant pods, and number of seeds per pod. These results are similar to those obtained in our study. Results recorded in this study may contribute to the development of an effective method to select components for heterosis breeding, combining elements of both quantitative and molecular genetics.

5. Conclusions

Genetic variability determines the values of heterosis effects in a very statistically significant way. This influence is directly proportional to the absolute value of the heterosis effect. Phenotypic diversity also affects the values of the heterosis effect. Phenotypic variation has a significant impact on SCA values. The results obtained can be a very valuable source of information for breeders in the perspective of conducting selection breeding. As the best parental forms recommended for further inclusion in breeding programs, preference should be given to those with the highest possible genetic variability, with relatively low phenotypic variability.