Genetical analysis for wheat genotypes using a half-diallel model under different sowing conditions

Wheat is a very important and economically significant crop worldwide, with a cultivated area of 222 million hectares, producing 779 million metric tons of grain in the 2021–2022 season¹. The yield production is suffering from many environmental factors such as drought^2,3,4,5,6, salinity^7,8,9,10, plant pathogens^11,12,13, and heat stress¹⁴. Increasing temperature and consequent changes in climate adversely affect plant growth and development, resulting in catastrophic loss of wheat productivity. For each degree rise in temperature, wheat production is estimated to be reduced by 6%; Elevated temperatures adversely affect various physiological, biological, and biochemical processes in wheat; Heat stress impacts seed germination, grain filling duration, grain number, Rubisco enzyme activity, photosynthetic capacity, assimilate translocation rate, leaf senescence, chlorophyll content, and overall yield¹⁵. Climate changes like increasing temperatures and declining rainfall have been considered significant reasons for declining wheat productivity especially due to the shortened grain filling period leading to reduction in grain size for wheat¹⁶. The late planting of wheat significantly affects spike plant⁻¹, spikelets spike⁻¹, and grain weight which eventually results in decreased yield causing a loss of 38% in the grain yield¹⁷. As a crop affected by temperature, late-sown crops are subjected to low temperatures during establishment and high temperatures during the reproductive phase, resulting in rapid crop maturity. This condition may harmfully impact on yield and yield components of wheat and finally reduced kernel weight¹⁸. Heterosis is the ability of a hybrid to outperform the mid-parent value or better parents. Under the influence of a fickle environment, the gene’s allelic or non-allelic relationships show hybrid vigor, in wheat, hybrid vigor has a direct association with the effective selection of potential parents¹⁹. The combining ability analysis is considered the most common biometrical tool to search out the parental genotypes from their potential to combine in hybrids²⁰. It divides the genetic variation into general combining ability, which measures the additive gene effects, and specific combining ability (SCA), which measures the non-additive gene action. Plant breeders could use these results for types of genetic variants in the traits for which selection is intended, as well as estimation of yielding ability by finding crosses that will generate superior genotypes for sowing date. These methods of adaptation could be suitable under changing climate conditions, particularly terminal heat stress. Thus, this study aimed to estimate the combining ability effects under normal and late planting dates, besides evaluating the heterotic amounts for the yield and yield components characters under normal and late sowing dates to use this information in the wheat breeding program.

Six genotypes for bread wheat were selected as parents in this investigation, representing a wide range of diversity for several earliness and agronomic characters. Table 1 contains the names of the parents, as well as their pedigree and origin. During the 2021/2022 season, parental genotypes were sown at two various dates in order to overcome the differences in flowering time. All possible parental combinations excluding reciprocals were made among the six genotypes, giving 15 crosses.

In 2022/2023 season, the 21 entries (6 parents and 15 F₁) were examined in 2 distinct planting dates tests. The first experiment was sown on the optimum planting date of 12th November while, the second experiment was sown on late planting date of 14th December. Due to climate changes and rising temperatures in the case of late wheat plantings after late summer crops such as peas and potatoes, wheat plants are subjected to high heat stress at the end of the season during the flowering and ripening period.

During soil preparation, both 2 investigations received the fertilization with 15 kg P₂O₅/fad, 24 kg K₂O/fad in single dose, and 75 kg N/fad supplied in two equal doses. Following 27 days from planting, the first dose was 30% with planting and the second was 70% with the first irrigation. Two experiments were carried out in the experimental farm, faculty of agriculture, Mansoura University, Dakahlia Governorate, Egypt, utilizing randomized complete block design (RCBD) with three replications. Every replication had 21 rows (genotypes) beside two rows (borders) that were 4 m long and 25 cm apart, with a 20 cm spacing among plants. Each row was sown with twenty grains, which were then manually drilled. All other cultural techniques, except for planting dates, were followed as indicated for wheat cultivation. To reduce border impact, removed the two outside plants and the two exteriors of each row in each plot for each character, ten plants were used to determine the attributes. The studied characters were number of spikes plant⁻¹ (SP⁻¹), number of tillers plant⁻¹ (TP⁻¹), number of spikelets spike⁻¹ (SPS⁻¹), number of kernels spike⁻¹ (KS⁻¹), spike length (SL, cm), kernels weight spike⁻¹ (KWS⁻¹), 100-kernels weight (100KW, g) and grain yield plant⁻¹ (GYP⁻¹, g).

Plot mean analysis was performed on the collected data for each trait in the parents and F₁. To examine the association between genotypes and sowing date conditions, all collected data were analyzed statistically using the randomized complete block design method, as explained by Gomez and Gomez²¹. A combined analysis of the two investigations was carried out (optimum and late sowing dates) to indicate the sowing dates effects according to Snedecor and Cochran²². The Griffing method 2 model 1 was utilized in an ordinary analysis to measure the impacts of general combining ability (GCA) and specific combining ability (SCA: For individual crosses, heterosis was defined as the percentage difference between the F₁ means and the mid-parent means (MP) and better parent (BP), as defined by Mather and Jinks²³.

Analysis of variance for the yield and yield components characters is presented in Tables 2 and 3. The results showed that the mean of squares of genotypes, parents, and their F₁ were P < 0.05 or P < 0.01 for most studied characters under both sowing dates_. However, parents versus crosses mean squares as an indication to average heterosis overall crosses were P < 0.05 or P < 0.01 for spike length, kernels weight spike⁻¹, 100-kernel weight under normal sowing date.

Analysis of variance of the combined analysis for the yield and yield components are presented in Tables 4 and 5. The results indicated that sowing dates mean squares were P < 0.05 or P < 0.01) for all traits except for grain yield plant⁻¹. The genotype’s mean squares were P < 0.05 or P < 0.01) for all the studied traits. The sowing dates × genotypes interactions were P < 0.05 or P < 0.01 for spikes plant⁻¹, tillers plant⁻¹, spikelets spike⁻¹ and kernels weight spike⁻¹. These results agree with those obtained by Aboshosha et al.²⁴, Shaban et al.²⁵ and Parveen et al.²⁶.

Combining ability

Analysis of combining ability variance

Mean squares of general and specific combining ability were P < 0.01 for all traits under normal and late sowing dates except for SP⁻¹, TP⁻¹, 100-KW and GYP⁻¹ under late sowing date due to GCA; while for SP⁻¹, TP⁻¹, SL, 100-KW under late sowing date, and KS⁻¹, KWS⁻¹, and GYP⁻¹ under normal and late sowing dates due to SCA, as presented in Tables 6 and 7. The significance of GCA and SCA indicates the presence of both additive and non-additive types of genes in the genetic system controlling these traits. Both GCA and SCA variances were P < 0.01 for most traits under both sowing dates, indicating the importance of additive and non-additive effects in determining the performance of these characters. The GCA were higher than those of SCA for most traits under the study except for, SP⁻¹, TP⁻¹ and 100-KW under normal and late sowing date. The results obtained revealed that the ratio of GCA/SCA was more than unity for SPS⁻¹, KS⁻¹, SL, and GYP⁻¹ under normal and late sowing dates and KWS⁻¹ under normal sowing dates.

These results suggest that the studied traits are predominantly controlled by additive gene action. It could be concluded that selection would be successful in improving these traits and selection would be more effective in the early segregated generation. On the other hand, the ratio of GCA/SCA was less than unity for SP⁻¹ and 100-KW under normal and late sowing dates, and KWS⁻¹ under late sowing date, therefore, selection would be more effective in the late segregated generation. These findings agree with those obtained by Aboshosha et al.²⁴, Hassan et al.²⁷, and Tolwani and Shulka²⁸.

General combining ability effects (GCA)

The estimates of general combining ability effects of the parental genotypes for yield and yield components characters at normal and late sowing dates are presented in Tables 8 and 9. Significantly positive GCA values would be the best general combiner for all studied characters where would be useful from the breeder point of view. Based on general combining it could be concluded that the best combiners for spikes plant⁻¹ were Sakha 94 and Sakha 95 under normal sowing date; for TP⁻¹ was Sakha 95 under the normal sowing date; for SPS⁻¹ was Sakha 94 under normal sowing date and Sakha 95 under normal and late sowing dates; for KS⁻¹ were Sids 14 and Sakha 94 under normal sowing date, and Giza 168 under late sowing date. However, for SL were Sids14 and Giza168 under both dates; for KWS⁻¹ was Sids 14 under both dates; for 100-KW was Line1 under the normal sowing date. On the other side for GYP⁻¹ was Sakha 95 at both dates.

Specific combining ability effects (SCA)

The estimates of specific combining ability effects of the F₁ crosses were computed for yield and yield component characters at normal and late sowing dates are presented in Tables 10 and 11. Significantly positive SCA values would be the best cross combinations for all studied characters, which would be useful from the breeder point of view. Based on specific combining ability estimates, it could be concluded that the best crosses for spikes plant⁻¹ was Sakha95 × Giza168 (P₃ × P₄) under normal sowing date, Sids 14 × Sakha 95(P₁ × P₅) and Line1 × Misr3 (P₅ × P₆) at late sowing date; for TP⁻¹ was Sakha 95 × Giza168 (P₃ × P₄) under normal sowing date; for SPS⁻¹ Sakha95 × Giza168 (P₃ × P₄) under normal sowing date and Giza 168 × Misr 3 (P₄ × P₆) under late sowing date; for KS⁻¹ Sids 14 × Line 1 (P₁ × P₅) under both sowing dates. However, for spike length Sids14 × Sakha 95 (P₁ × P₃) under normal sowing date; for KWS⁻¹was Sakha 95 × Line1 (P₃ × P₅) under late sowing date, respectively for 100-KW; Meanwhile, for GYP⁻¹ were Sids 14 × Sakha 95(P₁ × P₅), Sakha 94 × Giza168 (P₂ × P₄) at normal sowing date and Line1 × Misr3 (P₅ × P₆) at both sowing dates. These results agree with those obtained by Kamara et al.²⁹, Riaz et al.³⁰, and Ahmed and Gupta³¹.

Heterosis percentage

The estimations of heterosis over mid and better parents for yield and yield components characters under normal and late sowing dates are presented in Figs. 1, 2, 3, 4, 5, 6, 7 and 8. Significant or highly significant positive heterosis values over the mid and better parents would be the best for all the characters under normal and late sowing dates. Significant positive heterosis values for SP⁻¹ and TP⁻¹ were obtained by the cross (P₂ × P₃) over their mid parent under normal sowing date and the cross (P₅ × P₆) for SP⁻¹ under late sowing date over the better parent; while highly significant positive heterosis values for SPS⁻¹ and SL (P₂ × P₃) and (P₂ × P₄) over their mid and better parent under normal and late sowing dates; while, for KWS⁻¹ highly significant positive heterosis values over the mid parent were obtained at the most cases under both sowing dates. However, highly significant positive heterosis values over the better parent were obtained by (P₂ × P₄), (P₂ × P₅), (P₂ × P₆), (P₃ × P₄), and (P₃ × P₆) under both sowing dates and under late sowing date for 100-KW highly significant positive heterosis values over their mid and better parent were obtained in most cases (Figs. 6 and 7). For grain yield, results indicated that positive heterosis values over their mid-parent were obtained by crosses P1 × P6 (9.34%) and P5 × P6 (20.45%) under normal sowing date (Fig. 8). On the other hand, positive heterosis values over their mid-parent were obtained by crosses P1 × P5 (7.73%), P4 × P6 (12.38%) and P5 × P6 (20.78%) under a late sowing date. While positive and the highest percentages of heterosis over their better parent were obtained by crosse P1 × P6 (9.34%), P2 × P4 (7.09%), and P5 × P6 (7.8%) under normal sowing date and late sowing date. Heterosis is a complex genetically phenomenon that depends on the balance of different combinations of gene effects as well as the distribution of plus and minus alleles in the parents of a mating. So, heterosis is considered as the best tool to increase or break the yield barriers³². The amount of heterosis is the difference between the crossbred and inbred means³³. Wheat shows hybrid vigor when hybridization occurs between varieties. Hybrid vigor is based on the theory of dominance, which states that dominant alleles are responsible for vigorous growth and increased yield, or these dominant alleles prevent the harmful effects of recessive alleles by dominating them. Or perhaps it is based on the theory of heterozygosity, i.e. physiological stimulation is the basis of this theory, assuming that genetic mixing creates a type of physiological stimulation in the organism, resulting in the hybrid individual’s traits being superior to any purebred individual. Genetic mixing is the basis of hybrid vigor, therefore, hybrid vigor cannot be established in a purebred lineage. This theory is also called the theory of overdominance, due to the interaction of hybrid alleles and their superiority over purebred parents. These results agree with those obtained by Baloch et al.¹⁹, Aboshosha et al.²⁴, Abd El-kader et al.³⁴.

Our results revealed that the variance of sowing dates was P < 0.05 or P < 0.01 for all traits except for grain yield plant⁻¹. The variance of general (GCA) and specific (SCA) combining ability were P < 0.05 for examined characteristics at each sowing date except for spikes plant⁻¹, tillers plant⁻¹, 100-kernel weight, and grain yield under late sowing date due to GCA; whereas for tillers plant⁻¹, spike length, 100 kernels weight under late sowing date, and kernels spike⁻¹, and grain yield under normal and late sowing dates due to SCA. GCA/SCA ratio was more than unity for kernels spike⁻¹, spike length, and grain yield plant⁻¹ under normal and late sowing dates and kernels weight spike⁻¹ under normal sowing dates. These results indicated that the control of these characteristics was more dependent on additive genetic impact. Our findings indicated that the parent Sakha 95 was the best combiner for grain yield, while, Line 1 × Misr 3 was the best for grain yield based on Specific combining ability. Therefore, our results recommend use of these hybrids in subsequent successive studies until obtaining pure lines in the fifth generation, which may be considered a miniature model for a program to produce lines that are characterized by early maturity and high productivity with excellent quality traits, in addition to their tolerance to changing climatic conditions, especially in the case of late planting. Such advancements could be considered a step toward narrowing the gap between bread wheat production and consumption.