Based on the results of the analysis of variance, genotypes for all traits showed significant differences in two years. Significant differences were observed in sowing dates for all traits in two years. Also, cultivar×place interaction for most traits in the first and second years showed a significant difference (Tables 1 and 2). In other words, environmental factors had a significant effect on genotypes, which ultimately affected the phenological stages. On both sowing dates, during two years of experiments, the average of all phenological traits was between two parents Kauz (heat tolerant) and Montana (heat sensitive). However, the extreme values exceed the parental genotypes, which indicates that the genetic transgressive segregation for phenological traits has been well established and was maintained in proceeding generations, as illustrated by Koide et al. (2019). Concurrently, an appropriate diversity has been created within the RILs population for phenological features (Fig. 2). The existence of differentiation shows that the alleles of both parents have been effective in expressing the phenotype (Talukder et al. 2014). The observed transgressive segregation in the present population can be due to the simultaneous segregation of the genes controlling the studied traits and the accumulation of favorable alleles in the individuals within the population. It is so helpful to find the genetic variation for improving the production of plant species under different conditions (Khodadadi et al. 2011). The results showed that growth duration in SY2 (110 days) decreased compared to NY2 (149 days). In NY2 TI, SE, BO, HD, ANT, MD, DOU, GFD and DPM stages decreased by 42, 59, 95, 102, 109, 116, 135, 40 and 149 days to 48, 54, 72, 77, 82, 87, 101, 29 and 110 days at SY2, respectively. Decrease growth duration of phenological traits in the first year was also remarkably observed so that the total growth period decreased from 154 days from NY1 to 94 days at SY1 (Fig. 2). Sah et al. (2016) stated that plants use different mechanisms to tolerate environmental factors, including reducing their developmental period.
Table 1
ANOVA for phonological traits, with their coefficient of variation and broad sense heritability in first year
SOV | df | Mean squares |
TI | SE | BO | HD | ANT | MD | DOU | DPM | GFD |
Place | 1 | 863*** | 26447*** | 193508*** | 238669*** | 297013*** | 470881*** | 535278*** | 557523*** | 132800*** |
Rep(Place) | 2 | 212 | 1085.2 | 4915.8 | 1744 | 2559 | 3141 | 1539 | 201 | 1288 |
Genotype | 155 | 28*** | 139.7*** | 580.8*** | 685*** | 575*** | 341*** | 207*** | 105* | 251*** |
Genotype × Place | 155 | 19ns | 62.9ns | 369*** | 366*** | 369*** | 99*** | 105*** | 108* | 320*** |
Error | 310 | 19 | 52.15 | 87.9 | 44.19 | 64.8 | 58 | 62 | 79.6 | 81 |
CV | - | 13.58 | 7.22 | 9.37 | 6.64 | 8.5 | 7.35 | 6.9 | 7.2 | 18.5 |
h2 | - | 66.27 | 75.85 | 73.77 | 77.93 | 74.13 | 84.2 | 75.27 | 58.69 | 58.2 |
*P < 0.05; **P < 0.01; ***P < 0.001; ns: not significant. |
Table 2
ANOVA for phonological traits, with their coefficient of variation and broad sense heritability in second year
SOV | df | TI | SE | BO | HD | ANT | MD | DOU | DPM | GFD |
Place | 1 | 4326.9*** | 3326.19*** | 84364.7*** | 100810*** | 114492*** | 130012*** | 166987*** | 224036*** | 20963*** |
Rep(Place) | 2 | 1504.97 | 85.27 | 28.23 | 127.93 | 81.45 | 108.47 | 75.90 | 33.14 | 79.03 |
Genotype | 155 | 80.96*** | 126.94*** | 666.44*** | 822.89*** | 831.94*** | 809.17*** | 552.39*** | 306.37*** | 197.08*** |
Genotype × Place | 155 | 80.71*** | 56.98*** | 169.35*** | 188.19*** | 179.33*** | 170.12*** | 182.99*** | 122.50*** | 54.56 ** |
Error | 310 | 42.93 | 16.79 | 57.90 | 25.56 | 20.02 | 21.26 | 22.80 | 51.09 | 51.55 |
CV | - | 14.45 | 7.25 | 9.10 | 5.69 | 4.73 | 4.56 | 4.06 | 5.54 | 7.6 |
h2 | - | 0.61 | 0.80 | 0.87 | 0.89 | 0.90 | 0.90 | 0.85 | 0.81 | 0.83 |
*P < 0.05; **P < 0.01; ***P < 0.001; ns: not significant. |
Heritability of phenological traits in the second year was calculated in the range of h2 = 0.55–0.90. The highest values for ANT and MD were equal to h2 = 0.90. During the first year, similar to the second year, heritability range (h2 = 0.58–0.84) was calculated (Table 1 and Table 2). The presence of high heritability indicates the greater effect of genes compared to environmental factors in the expression of traits, which indicates the accuracy and optimal conditions for QTL detection.
The description of QTLs detected in present study are presented in supplementary data file 1. Among the 93 QTLs located in this study, some QTLs for TI were found on chromosomes 2D, 2A, 5B, and 3B; the highest R2 = 12.86% and 11.64%, which explained by TI3.Y1S-2D and TI1.Y1N-3B QTLs on chromosomes 2D and 3B, respectively. The identified QTLs for this trait were stable only in a sowing date, one year, and as minor QTL. QTLs of SE phase were varied on the chromosomes of all three A, B, and D genomes, and were located on chromosomes 2D, 5A, 1B, 4D, 5B, and 7B. The distance between two gwm132-gwm484 markers contained two major QSE2.Y2N-2D and QSE2.Y2S-2D QTLs in 18.83 and 26.83 cM points, respectively. These two QTLs explained R2 = 17.48% and 20.79% of phenotypic variance in NY2 and SY2, respectively. In different regions of chromosome 5A in three environments except for NY2, three QTLs were located as minor and major QTLs for SE between gwm126-gwm291 markers. Based on the climatic conditions, various QTLs of BO were founded on chromosomes 1A, 1B, 2D, 3D, 4D, 5A and 5B. The BO3Y2N-2D QTL on chromosome 2D in NY2 explained the highest amount of phenotypic variation (R2 = 55.98%), while other QTLs showed a low amount of phenotypic variation and were not stable in different sowing dates. Also, mapped QTLs of HD were located in different regions on different chromosomes but were not stable in any of the sowing dates. The QTL HD3Y2N-2D at 22.83 cM on chromosome 2D in NY2 with LOD = 14.6, explained 76.69% of phenotypic variation for HD trait. In NY2 and SY2, a QTL (ANT3Y2N-2D and ANT3Y2S-2D) at 22.83 cM for ANT could express 76.29% and 60.13% of phenotypic variation, respectively. Also, for ANT and HD in NY1, two QTLs were mapped on chromosomes 5A, both of which were minor and common for both traits. In SY1, the HD QTLs were located on chromosomes 2A and 1B, while ANT was found on chromosomes 6D and 7D (Fig. 3 and S1). MD, DOU, DPM and GFD QTLs were mapped the same in NY2 and SY2, with LOD = 2.3–18.5 and R2 = 7-58.44% at NY2 and LOD = 2.13–7.24 and R2 = 8-58.31% at SY2. Also, the most phenotypic variance was explained by QTL located at 22.83 cM in both NY1 and NY2. Except for QTL (DPM1.Y1S-4B) with a minor effect that was located on chromosome 4B at SY1, no QTL was located on this chromosome (Fig. 3). The results showed that in the second year for BO, HD, ANT, MD, DOU, and GFD traits, a common QTL was identified between two gwm750-barc17 markers at 0.01cM on chromosome 1A (R2 = 7-13.28%, at NY2 and R2 = 11.84–21.89% at SY2). The notable point is that at NY2, this QTL was identified as minor and at SY2 as QTL with major effect.