Multivariate analysis of genetic divergence among winged bean (Psophocarpus tetragonolobus L.) genotypes for effective hybridization programme

Introduction

Winged bean (Psophocarpus tetragonolobus L.) is a highly versatile yet underutilized food legume native to the humid equatorial tropics of Southeast Asia (Hymowitz et al., 1977) [3] . It ranks among the most nutritionally complete leguminous crops and forms a valuable component of traditional agricultural systems across tropical regions. It is a self-pollinating diploid species having a somatic chromosome number of 2n=2x=18 (Harder and Smartt, 1992; Vatanparast et al., 2016) [2,12] . The crop is well-adapted to warm and humid tropical climatic conditions and sustains productivity in low-input farming systems across countries including India, Papua New Guinea, Myanmar and the Philippines (Singh et al., 2022) [10] . Interestingly, all plant parts seeds, pods, leaves, flowers and tuberous roots are edible and nutritionally rich, which forced the National Academy of Sciences (1981) [7] to describe it as "a supermarket on a stalk". Its notable protein content, atmospheric nitrogen-fixing capacity and broad adaptability position it as a highly promising alternative crop for food and nutritional security in tropical developing nations (Tanzi et al., 2019) [11] . Genetic improvement of Winged bean begins with judicious selection of parental lines and the establishment of segregating populations from which superior genotypes are identified. A central challenge for breeders is how to choose appropriate parents for hybridization. Crosses between genetically divergent parents are known to generate greater heterosis than those between closely related genotypes. Genetic divergence analysis is therefore a vital tool for evaluating germplasm collections and guiding effective parent selection in crop improvement programmes. The present investigation was undertaken to estimate genetic divergence (D²) among Winged bean genotypes for identification of suitable parental lines for hybridization.

Materials and Methods

A total of twenty-eight genotypes were evaluated during the agricultural season 2024-25 at main experiment station, Department of Vegetable Science, College of Horticulture and Forestry, ANDUAT, Ayodhya, Uttar Pradesh, India. The experiment was laid out in a randomized complete block design comprising three replications with a plant spacing maintained at 125 × 60 cm.

All recommended agronomic and crop management practices were followed in a timely manner throughout the growing period. From each replication of every genotype five plants were selected at random, tagged and utilized for data collection. A total of nineteen morphological and yield-related characters were recorded as observations. To assess the extent of genetic divergence among the genotypes, Mahalanobis's generalized distance (D²) statistic was employed and the genotypes were subsequently categorized into distinct clusters using Tocher's method.

Results and Discussions

Analysis to estimate D2 value was done on the basis of relative magnitude of D2 values, all the twenty eight genotypes were grouped into 4 clusters (Table 1). Among the clusters, cluster III was the biggest cluster with 11 genotypes. Cluster I was the second largest cluster with ten genotypes followed by cluster II with six genotypes. The cluster IV were consisting of only one genotype. Similar patterns were also observed by Prasanth et al. (2016) [8] and Kashyap (2021) [5] . The inter and intra cluster distances among the genotypes have been shown in table 2. Among all inter-cluster distances recorded the maximum was noted between cluster II and cluster IV (14.203), which was subsequently followed by the distance between cluster III and cluster IV (12.413) and then between cluster I and cluster IV (11.194). Regarding intra-cluster distances the minimum value excluding the solitary cluster was recorded in cluster II (2.640). Cluster IV exhibited an intracluster distance approaching nearly zero. Two intermediate clusters falling between the minimum and maximum values were also identified, representing distinct divergence groups. On the other hand, the maximum intra-cluster distance was recorded in cluster I (3.159), with cluster III (2.895) following closely behind it. Roy (2021) [9] and Gadi et al. (2025) [1] also reported similar patterns of variable inter and intra-cluster D² distances. With reference to the cluster wise mean values presented in Table 3, the mean pod yield/hectare and seed yield/hectare were found to be the highest in cluster IV (167.74 q and 49.5 q, respectively), whereas the lowest mean values for both these traits were recorded in cluster II (30.97 q and 5.88 q, respectively). Among the remaining clusters, cluster I registered the highest mean value for total soluble solids (5.38), while cluster II did not record the highest mean value for any of the traits under study. The highest mean value for number of flowers/inflorescence (10.20) was exhibited by cluster III. Meanwhile, cluster IV which was represented by a single genotype demonstrated the highest mean values across all the remaining traits which included number of seeds/pod (18.37), seed index (40.70 g), pod length (28.33 cm) and others. The similar pattern of cluster-specific superiority for different individual traits, where different clusters recorded maximum mean values for different characters was observed by Mishra (2019) [6] . In the current study, the percentage contribution of various key traits toward genetic divergence was calculated and has been presented in Table 4. Among all the characters evaluated for their contribution to genetic divergence internode length recorded the highest contribution of 7.70%, followed by pod width (6.87%) and leaflet width (6.15%). Similar findings were also observed by Prasanth et al. (2016) [8] , who reported that certain morphological traits accounted for a higher share of divergence than others. Similarly documented by Kant et al. (2022) [4] and Kashyap (2021) [5] . Dry matter (%) 6.02 16.

Clusters

Green pod yield/plant (kg) 3.29 17.

Seed yield/plant (kg) 3.97 18.

Green pod yield/hectare (q) 4.72 19.

Seed yield/hectare (q) 4.75

Conclusion

The D² analysis grouped twenty-eight genotypes into four clusters revealing considerable genetic divergence. Crosses between genotypes from clusters II and IV exhibiting maximum inter-cluster distance are recommended for hybridization to obtain superior segregants. Internode length, pod width and leaflet width were the major contributors to genetic divergence.

Acknowledgement

AcknowledgementThe authors express sincere gratitude to the Department of Vegetable Science, College of Horticulture and Forestry, Acharya Narendra Deva University of Agriculture and Technology, Ayodhya, Uttar Pradesh, India, for providing all necessary facilities, resources and experimental infrastructure required to conduct the present investigation.

References

1. Gadi J, Seyie K, Shah P, Imam Z. Genetic diversity assessment of Winged bean (Psophocarpus tetragonolobus L.) genotypes. Journal of Advances in Biology & Biotechnology. 2025;28(2):673-680.
2. Harder DK, Smartt J. Further evidence on the origin of the cultivated Winged bean, Psophocarpus tetragonolobus (L.) DC. (Fabaceae): Chromosome numbers and the presence of a host-specific fungus. Economic Botany. 1992;46(2):187191.
3. Hymowitz T, Boyd J. Origin, ethnobotany and agricultural potential of the winged bean: Psophocarpus tetragonolobus. Economic Botany. 1977;31(2):180-188.
4. Kant A, Saha S, Sarkar S, Lagoriya DS, Paul SK, Nandan R. D2 Statistical Analysis of Tuber Yield and its Contributing Traits in Winged Bean [Psophocarpus tetragonolobus (L.) DC.]. Biological Forum -An International Journal. 2022;14(2):1010-1015.
5. Kashyap A. Study of genetic variability, character association and genetic diversity analysis for quantitative traits in Winged Bean (Psophocarpus tetragonolobus L.). Raipur: Indira Gandhi Krishi Vishwavidyalaya; 2021
6. Mishra A. Genetic analysis for yield and its components in Winged bean (Psophocarpus tetragonolobus L.). Raipur: Indira Gandhi Krishi Vishwavidyalaya; 2019
7. National Academy of Sciences. Winged Bean: A HighProtein Crop for the Tropics. Washington, DC: The National Academies Press; 1981
8. Prasanth K, Sreelathakumary I, Celine VA. Genetic divergence studies in Winged Bean (Psophocarpus tetragonolobus (L.) DC.). Indian Journal of plant genetic resources. 2016;30(1):84-87.
9. Singh PK, Tiwari JK, Joshi V, Lal SK, Selvakumar R, Kumar J, Kumar A, Mani I. Winged bean -A nutritionally rich underutilized vegetable crop. Indian Horticulture. 2022;67(1):16-19.
10. Tanzi AS, Eagleton GE, Ho WK, Wong QN, Mayes S, Massawe F. Winged bean (Psophocarpus tetragonolobus (L.) DC.) for food and nutritional security: synthesis of past research and future direction. Planta. 2019;250:911-931.
11. Vatanparast M, Shetty P, Chopra R, Doyle JJ, Sathyanarayana N, Egan AN. Transcriptome sequencing and marker development in winged bean (Psophocarpus tetragonolobus; Leguminoceae). Scientific reports. 2016;6(1):29070.