Assessment of genetic diversity by phenological traits, field performance, and Start Codon Targeted (SCoT) polymorphism marker of seventeen soybean genotypes (Glycine max L.)

Plant materials

Seventeen Egyptian soybean (Glycine max L.) genotypes were kindly supplied from the Genetic Resources Research Department, Field Crops Research Institute, Agricultural Research Center (ARC), Giza, Egypt. The pedigree and origin of genotypes were taken from Akram et al. (2011); Morsy, Mohamed & Abou-Sin (2016); Guo et al. (2022) and coded (Table 1). Planted place in Spring, 15th of May which is the soybean growing season, as soybean planting begins in May in Egypt.

Determination of phenological characteristics

The current experiment was carried out in the laboratory of the Agricultural Botany Department, Faculty of Agriculture, Al-Azhar University Assiut branch in the 2023 season to examine the phenological characteristics of soybean plant seedlings. The seeds of seventeen soybean genotypes were planted in 14 cm diameter plastic pots that were filled with 2 kg of a mixture of clay and sand (1:2 w: w), with 10–15 seeds in each pot. The seventeen cultivars in this experiment were divided into three replications (three pots for each replicate) Different phenological characteristics of plant samples were taken 26 days after sowing. Three plants per replicate from each genotype were randomly taken. The plants were separated into their organs (roots, stems, and leaves). Scan images from different genotypes were utilized to get quantitative measurements of the phenological traits as follows: Root traits (root number (RN), root maximum length (RL), root width (RW) and root tip angle (TA^∘)); Shoot traits (shoot length (SL), shoot length of first internodes (LFI) and shoot diameter of first internodes (DFI)); Leaf traits (leaf area (LA), leaf length (LL) and leaf width (LW)) and Seed traits (single seed weight (SW), seed length (SL), seed area (SA) and seed weight (SW)).

Field experiments

The current experiment was conducted over two years 2022/2023 at Bahtim Research Station, Agricultural Research Center, Kaliobia Governorate, Egypt (Latitude 30°8′22″N, Longitude N 31°15′50″) to examine the field performance of soybean crops. The seeds of seventeen soybean genotypes were planted after where one waits until the soil is firm and can withstand the feet, and then the row is opened and planted. Planting was in rows in three replicates, each replicate consisting of three rows each 3 m, and the distance between the rows is 70 in a randomized complete block design (RCBD), and the distance between the holes is 15 cm. The soybean genotypes were planted at a density of 30 plants/m in a single row on the row. Soybean genotypes received the standard agricultural practices according to the recommendation of the Ministry of Agriculture, Egypt by instructions agriculture extension were followed regarding fertilization and irrigation until the harvest was complete. The experimental soil texture was clay. The soil physicochemical properties were characterized by ana-lysing samples from 30 cm depth (Table 2). Growth parameters including plant height, number of branches, number of pods, number of seeds, seed yield, and one hundred seed weight were measured.

Molecular evaluations

To examine the SCoT polymorphism marker, approximately 500 mg of young and fresh leaves from each genotype of five-week-old plants of seventeen soybean genotypes were collected from the first experiment. Genomic DNA was extracted from fresh plant leaves by the DNeasy plant mini kit 69204 (Bio Basic, Amherst, MA, USA). To assess DNA purity, the ratio of absorbance at 260 and 280 nm is used using a UV spectrophotometer. The consensus sequence was used to design ten SCoT primers from Joshi et al. (1997); Collard & Mackill (2009); Mohamed, Shoaib & Gadalla (2015); Alotaibi & Abd-Elgawad (2022). All SCoT primers were 18-mer and were from Dataset I, which is based on highly expressed genes as described by Sawant et al. (1999). Eight primers were selected, which gave very noticeable and consistent bands for the data’s final amplification (Table 3). Amplification reactions were performed following (Xiong et al., 2011; Fathi, Hussein & Mohamed, 2013; Alotaibi & Abd-Elgawad, 2022) where, an annealing at 50 °C for 50 s. and a prolongation stage at 60 s. The preliminary expansion section was expanded to 7 min at 72 °C within the last cycle and was carried out in the Techni^®TC-512 Thermal Cycler (UK). Amplified PCR fragments were electrophoresed on 1.5% agarose gel with staining dye ethidium bromide to visualize DNA and a 1kb ladder marker (Bio-Rad, Hercules, CA, USA). The runs were carried out for exactly 30 min at 100 V in a mini-submarine gel (Bio-Rad).

Statistical analysis

All phenotypic data were evaluated for normal significance using the Shapiro–Wilk test (α = 0.05), the standard deviation was calculated, and a one-way ANOVA using SPSS ver. 22.0 and a graphical representation was created using a JMP^® ver.16 cell plot (SAS Institute Inc, 2008 software version 9.13). To compare a genotype’s performance, PLABSTAT software (Utz, 2001) used genotypes and the three replications as fixed and random effects. The following parameters were calculated with a 5% criterion of significance for all tests: Heritability (H2) = Genotypic variance (2G)/ Phenotypic variance (2p); Genetic coefficient of variation (GCV) and Correlation coefficients for each trait using PLABSTAT software’s GENOT function.

For phenotypic data, a constellation plot of the hierarchical clustering dendrogram was produced using JMP^® ver.16 and Ward’s technique. For field data, clustering of genotypes was performed based on Ward linkage and Euclidean distances as an “r” matrix (Everitt, 1993; Eisen et al., 1998).

DNA banding patterns were photographed using the Bio-1D gel documentation system. All gels were analyzed by Gel Analyzer 3 software, which scored clear amplicons as present (1) or absent (0) for each primer and entered them in the form of a binary data matrix. The following are several descriptive measures of diversity: Band’s number (total, monomorphic, polymorphic, and unique); percent of polymorphism (Pb %) and Marker efficiency indices (Heterozygosity index (H), Polymorphic Information Content (PIC), effective multiplex ratio (E), Arithmetic mean of H (H.av), Marker In-dex (MI) and discriminating power (D) according to Chňapek et al. (2024), calculated by the Marker Efficiency Calculator (iMEC, 2018)). The Dice coefficient was used to assess the genetic similarity of genotypes using a (0/1) data matrix (Adhikari et al., 2015; Khattab et al., 2022). Heatmap was performed by using the R software through the web tool ClustVis (Adhikari et al., 2015) to visualize similarities and dissimilarities among genotypes. Collectively, ClustVis, a web tool for visualizing the clustering of multivariate data was used to illustrate and calculate relationships, heatmap, and genetic trees based on penology, field performance traits, and SCoT according to Metsalu & Vilo (2015).

Phenological traits

In this study, fourteen quantitative traits for the seventeen soybean genotypes were examined. The mean root parameters and shoot stem parameters of the seventeen soybean genotypes are shown in Table 4. Root parameters and shoot stem parameters showed significant differences among all genotypes (P < 0.05). For root parameters (Table 4 and Fig. 1), the root number (RN) of all genotypes was between 6.67 and 26.67 while the shortest (6.67 cm) was recorded in H113. Root maximum length (21.66 cm) (RL) was recorded in H30 genotype while the shortest (6.91 cm) was recorded in H117. The root width (RW) (cm) of Crawford was the highest (0.40 cm) while the lowest (0.25 cm) was recorded in D75_12035. Ttip angle of root T.A of H117 was the longest (64.19°), while, the shortest (26.93°) was recorded in H113. Herein, the Crawford genotype had the highest root width value (0.40 cm) and the lowest (0.25 cm) was found in the D75_12035 genotype. Results illustrated that the largest root tip acute angle (TA) was in the H117 genotype (64.185°) and the smallest was recorded in the H113 genotype (26.928°). Additionally, in the H113 genotype, a wider root system (RW) was linked to a reduced degree of tip angle for seminal roots. For shoot stem parameters, the significant differences revealed that there is significant variability among the soybean genotypes for all the characteristics investigated at 0.05 level of probability (Table 4 and Fig. 1), shoot length (SL) (cm) of NC105 was the longest (21.83 cm), while, the shortest (9.10 cm) was recorded in H117. The length of the first internodes (LFI) values was between 4.96 and 7.65 cm. The diameter of the first internode (DFI) values ranged from 0.17–0.33 cm.

The mean values for the measured and calculated leaf parameters and seed parameters of the seventeen soybean genotypes are shown in Table 5. All genotypes demonstrated substantial differences in the leaf parameters and seed parameters phenotypic traits (P <  0.05). Phenological screening showed that the three essential leaf traits’ range: area (LA), length (LL), and width (LW) were in the ranges of 6.442–13.884 cm2, 2.603–4.256 cm, and 2.756–4.967 cm and, respectively. The data revealed that the widest leaf area was from in NC105 genotype while the smallest one for in the Toano genotype, respectively. Collectively, the Toano genotype had the lowest LL, LW, and LA values, while the NC105 genotype had the highest SL, LFI, and LA values, and also and H113 genotype had the maximum LL and LW values. Analysis of seed parameters of soybean genotypes are shown in Table 5 clearing the significant differences between means (P < 0.05). Single seed weight, seed length, seed area, and seed weight of the 17 soybean genotypes ranged from 7.21 to 9.29 9, 6.71 to 9.01 cm, 43.25 to 56.49 cm, and 0.28 to 0.61, respectively. All genotypes demonstrated substantial differences in the root phenotypic traits. In this study, modern automated and semi-automated methods of geometric shape analysis of phenotypic and geometric traits depart from the coordinates’ points for seed parameters. These traits are single seed weight (SW), seed length (SL), seed area (SA), and weight of seeds (WS). The genotypes NC105, DR101, and H129 had the highest values, whereas the genotypes DR101, Crawford, H117, and D75_12035 had the lowest values (Fig. 1 and Table 5).

The results of phenological traits showed tremendous variation among 26-day stage (26 days) soybean genotypes. These variations are essential for developing seventeen new cultivars with distinct phenological traits. The clustering between phenological and morphological characteristics is shown as a double dendrogram or colored heatmap (Fig. 1). The 17 genotypes were classified into two main clusters. Genotypes: NC105, H113, H129, H30, H32, DR101, Giza111, Crawford, Hutcheson belong to one group but H30 & H32, H113 & H129, Giza111 & Hutcheson more correlated with each other. The other group included H117, NC104, H2L12, H15L5, Toano, Holladay, and Giza35 D75_12035. The intensity of colors reflects visually the high, intermediate, and low characteristic values. Dark colors like blue usually indicate higher data values for genotypes NC105, H113, H30, and Crawford. Cooler colors like green and yellow represent lower data values which mostly for second group genotype Toano in leaf traits.

Field performance

Through the data across the two seasons, the data was collected and a homogeneity of variance analysis was performed in the 2022/2023 seasons via Bartlett’s test. For this, pooled variance across the two seasons was used. The combined analysis of variance after the homogeneity test for error variances, the F-test, and the mean performances of the seventeen soybean genotypes for different traits are performed (Table 6). The test revealed that there were no significant differences in the effect of the year for all traits, while the effect of genotypes was significant for all traits except for 100-seed weight. Analysis of the field behavior of the seventeen soybean genotypes are shown in Table 6. There were significant differences between all the field performance parameters of the 17 soybean genotypes (P < 0.05). The significant differences revealed that there is a significant variability among the genotypes for all the characteristics investigated. The H32 genotype was the tallest plant (95.0 cm), while had small pod numbers (37.88). The H113 genotype had the lowest number of seeds per plant (88.63) with the lightest seed yield (15.76 g). The H30 genotype was a short plant (36.38 cm) with the fewest number of branches per plant (2.375 cm). Although the NC104 genotype was a small plant (35.88 cm), it had the largest number of seeds per plant (204.0). The Giza111 genotype had the largest pod numbers per plant (74.38) and the most number of seeds per plant (202.5). Other genotypes showed different parameters: Holladay genotype had short plants (34.13 cm), H117 genotype had the most profuse plants (4.5 cm), N105 genotypes had the largest pod numbers per plant (75.0), H2L12 genotypes had the small pod numbers (35.63) and N104 genotype had the greatest seed yield per plant (35.42 g).

In the present analysis, the similarity levels of the 17 soybean genotypes were assessed based on seed yield and its related traits (Fig. 2). They were classified into six main groups (clusters) with high similarity of over 50% in almost all genotypes. The first cluster contains five genotypes with 67% similarity: NC 105, Holladay, Giza 111, Giza 35, and Hutcheson. The second cluster has four genotypes with 64.8% similarity: H117, H113, H2L12, and Crawford. Third shows the highest similarity% (83.96) between genotypes: H129, D 75-12035. The fourth cluster encloses NC 104, H 30 geno-types which had the highest pod numbers, seed yield, and 100 seed weight with 66.63% likeness. The fifth clustering includes the H32 genotype that has the highest number of plants and seed number/plant height with 40.67% similarity. Sixth clustering includes H32 and H15L5, DR101, and Toano by 77.42% similarity. The most noticeable relation between phenological characteristics with the field performance is the steady clustering of some genotypes among them than other genotypes: NC105, Giza 111, Hutchesonin & H113, H32, Crawford & H117, H2L12 & H15L5, Toano.

SCoT polymorphism marker

Out of the 15 primers used in this study, the eight SCoT primers that worked well produced good, repeatable, highly informative, and scorable fingerprint patterns (Fig. 3). Every SCoT primer displayed a distinct banding characteristic with a medium-to-high GC range of 50% to 67%. Out of 20 bands, 12 polymorphic bands were produced by the eight SCoT primers in this instance (Table 7). With an average polymorphism rate of 57%, the percentage of polymorphic bands varied from 50% with SCoT 2 to 75% with SCoT 3. The eight SCoT primers that were selected had varying degrees of effectiveness, which are listed in Table 7. With a mean moderate value of 0.30, the polymorphism information content (PIC) varied from 0.19 for SCoT 10 to 0.46 for SCoT 2. The Heterozygosity Index (H), Arithmetic Mean of H (H.av), Marker Index (MI), and Discriminating Power (D) were all greater for SCoT 3 and SCoT 9. Their mean values range from 0.35 to 0.40, and their values are H: 0.54, 0.52, H.av: 0.54, 0.52, MI: 0.54, 0.52, and D: 0.61, 0.57, respectively.

The data for SCoT is displayed in a heatmap where the grid/row represents a genotype and each column represents a band (Fig. 4). The color and intensity of the boxes are used to represent changes in similarities in DNA (not absolute values). The color scale from dark red (the least-used parts of a gird) as in five genotypes of cluster II (NC105, H117, H113, Toano, Holladay) to blue (the most popular parts of the gird) for the rest of the genotypes of cluster I. Dark red indicate more activity of genes, while cooler colors indicate less.

The complete assessment of the phenological traits, field performance, and SCoT polymorphism marker

The hierarchical cluster analysis (HCA) and its associated heatmap explored the differences among the genotypes (Fig. 5). Rows correspond to the studied Soybean genotypes, whereas columns correspond to different phenology, Field parameters, and SCoTs markers. Low numerical values are green colored, while high numerical values are colored rose (see the scale at the bottom left corner of the heat map). Briefly, the HCA-associated dendrogram based on seedling morphological characterization (phenology trait), field performance, and SCoT molecular markers outputs showed that they were clustered separately into four distinct clusters. Cluster ‘A’ included NC105, Giza 111, Hutcheson, and H30, while cluster ‘B’ consisted of NC104 Toano, Holiday, and Giza 35. Furthermore, cluster ‘C’ included H117, H29, H113, H32, D7512035 and H15L5, while H2L12, DR101, and Crawford were located in cluster ‘D’. Concerning the heatmap correlation, cluster A &B genotypes had a strong correlation. Moreover, there were moderate correlations among genotypes of cluster C, while the strongest correlations were found between cluster D genotypes.