GENETIC DIVERSITY AMONG SOURSOP GENOTYPES BASED ON FRUIT PRODUCTION

This study aimed to estimate the genetic diversity among soursop genotypes in terms of fruit yield evaluated in different crop years. Sixteen measurements for fruit yield in 71 soursop genotypes were carried out from 2000 to 2016. Based on ANOVA it was verified the existence of genetic variability among genotypes in the different measurements (harvests). The genotypes were clustered according to their respective fruit yield averages in the 16 years as well as from their means obtained in each measurement year by the Scott-Knott test (p<0.05). To study genetic diversity among genotypes, three methodologies were compared: Tocher hierarchical optimization, UPGMA method and principal components. Five groups were formed for all clustering methods used. It was identified that crossings between genotypes 124 and 145 with genotypes 59 and 170 are potential to generate population with large genetic variability and high fruit yield average. New researches should be developed aiming to exploit the genetic variability among soursop genotypes based on the results found in this study.


INTRODUCTION
The soursop (Annona muricata L.), belongs the Annonaceae family, and is one of the approximately 119 species that compose this genus Annona, which is widely distributed in different parts of the world.São josé et al. (2014) reported that the center of diversity of the species in this genus is widespread among Tropical America and Central Africa.Nevertheless, there is a controversy about the exact region of origin of the soursop, which according to Campos et al. (2008) is more probably that the origin was in the Central America, on the other hand FALCAO et al. (1982) considered that this species originated in the Antilles and spread in tropical America.
High diversity can be observed among species into the genus Annona, however only nine species produce edible fruits, and only five are cultivated with economic purpose (SÃO JOSÉ et al., 2014).Soursop stands out among the other species cultivated due to it presents fruits of considerable size, approximately 3 kg, which can be used for several purposes.Soursop pulp is commonly consumed fresh or can be frozen, processed and used industrially for the production of juices, ice cream, jellies, and another foods.However, the soursop production chain is largely composed of small producers and is often unsustainable because the fruits are highly perishable and lack adequate production and post-harvest techniques (FALCAO et al., 1982;SÃO JOSÉ et al., 2014).
Several parts of the plants can also be used in medicine for the production of herbal products standing out in the current market (MOGHADAMTOUSI et al., 2014).Increasingly, interest in studies in the field of medicine reinforces the need for greater knowledge and importance of soursop cultivation.Jaramillo et al. (2000) verified the action of their extracts against promastigotes of Leishmania braziliensis and L. panamensis, BORIES et al. (1991) tested its antiparasitic activity against Entamoeba.histolytica, Nippostrongylus.brasiliensis, and Artemia.salina, and several other authors reported the importance of this species in the treatment and prevention of important diseases, such as cancer (GEORGE et al., 2012;MOGHADAMTOUSI et al., 2014).
In this context, it is necessary to study soursop cultivation, aiming for increasing fruit yield, emphasizing the increase of nutritional quality, application of adequate techniques for seedling production, improvement of existing methodologies and efficient exploitation of genetic resources (CAMPOS et al., 2008;OLIVEIRA, 2001), to make the crop economically more attractive to farmers.
For the development and fixation of this crop in the market it is essential the availability of superior cultivars, with desirable traits such as yield (SANCHÉZ et al., 2017).Borém (2005) reported that 50% of the gain in production is related to cultural factors, present in the crop, while the other 50% are covered to the genetic attributes.However, there are few reports of research that investigated the stages of soursop genetic breeding (RONNING et al., 1995;SÁNCHEZ et al., 2017;SURATMAN et al., 2015).Among these stages, the initial phase characterized by the formation of the base population in which the selection will be practiced, pre-breeding studies are necessary to help breeders to identify superior genotypes to be used in the crossing blocks.Therefore, the identification of genotypes with desirable characteristics and with high diversity can contribute to generate segregating populations with high genetic variability and productive potential.
Therefore, the objective of this work was to estimate the genetic diversity among soursop genotypes in terms of fruit yield evaluated in different years, in order to obtain information to support future studies of soursop genetic breeding programs.

Experimental design
The experiment was laid out at campus Halcones Farm, in Cerritos -Risaralda, Colombia (4º48'48.00"N,75º42'58.63"W,at approximately 1,345 m asl).Region climate is AF according to Köppen's classification, with rainfall scattered throughout the year, with average temperatures of 23 ºC and minimum of 18 ºC, with annual rainfall of 2,100 mm and relative humidity around 75%.
The experiment was carried out using 8x8 m spacing.Plants were conducted without irrigation and the other cultural practices were carried out according to the recommended management practices for soursop (OLIVEIRA, 2001).Cultural practices were similar throughout the period of conduction of the experiment.
Fruit yield was evaluated in sixteen years from 2000 to 2016 in the 71 soursop genotypes.

Data analysis
Data were submitted to ANOVA, where the statistical model adopted considering two variation factors was: where Y ijk is the observation regarding the i th soursop genotype, in the j th measurement; µ is the overall mean; g i is the fixed effect of the i th genotype, under influence of the permanent environment; a j is the fixed effect of the temporary environment in the j th measurement, where ; and ijk ε is the experimental error established by the temporary effects of the environment in the j th measurement of the i th genotype.
The genotypes were clustered by the Scott-Knott test at a 5% probability according to their respective fruit yield means in the 16 years of evaluation, as well as from their means obtained in each measurement year.
Aiming to study the genetic diversity among genotypes, Tocher's optimization and UPGMA method, based on the standardized Euclidean mean distance, and principal components were applied.The criterion of Singh (1981) was used to evaluate the contribution of each measurement to the observed genetic diversity.Statistical data analyzes were performed using software Genes (CRUZ, 2013) and Microsoft Excel®.
Hierarquical clustering was also used to verify the diversity among soursop genotypes.UPGMA clustering based on the standard Euclidean distance allowed to visualize in a graphic how the genotypes were grouped (Figure 1).Two groups were formed, which the genotypes 124 and 145 belonged to group 1 and the remaining genotypes Genetic diversity… SÁNCHEZ, C. F. B. et al.
Biosci.J., Uberlândia, v. 34, n. 1, p. 122-128, Jan./Feb.2018 belonged to group 2. Therefore, it was possible to prove that the crossing between genotypes 124 and 145 with the genotypes 59 and 170 may be potential to improve the fruit yield in soursop because they belonged to different group by Tocher's optimization and UPGMA method, and they also presented a high fruit yield average during the years.
The genetic diversity should be explored aiming to choose genotypes with high general and specific combination ability that makes these genotypes useful to be crossed and generate hybrids with high heterosis.
Figure 1.Clustering of the 71 soursop genotypes using UPGMA method based on standard Euclidean distance.
The trait used to estimate standard Euclidean distance was fruit yield evaluated during 16 measurements.The experiment was laid out from 2000 to 2016.
Genetic diversity study has a big importance for the soursop breeding program success since this is one of the first step and it may help breeders to choose superior genotypes for crossings (CRUZ et al., 2012).Therefore genetic diversity study will help breeders to choose the best strategy based on their objectives (ALLARD, 1999).
Another methodology that was used in this study to calculate the genetic diversity among soursop genotypes was principal component analysis (Figure 2).This methodology is widely used in the genetic diversity study when there is no replication in the experiment, which means when there is just one plant represent each genotype.
To compare principal components analysis with Tocher's optimization, genotypes clustered in the same group by Tocher's optimization were represented using the same symbols, and for each group was used a different symbol (Table 3 and Figure 2).It was possible to verify that the groups formed by Tocher's optimization was similar for the groups formed by principal components analysis.This result is important because it demonstrated the precision between methods.
The maintenance of the genetic diversity is important for guarantee the possibility to obtain selection gain in a long term since variability is one of the most important effect that affects selection gain (RESENDE, 2002;RESENDE, 2015).Based on this fact it is necessary to cross soursop genotypes from different groups guaranteeing the variability on the soursop breeding program.Groups were established based on Tocher's optimization (each symbol represent one group based on Tocher's optimization -Table 3).The experiment was laid out from 2000 to 2016.
The criteria proposed by Singh (1981) was used to verify the contribution of each measurement for the genetic diversity among soursop genotypes, and it revealed that the 6 th , 7 th , and 8 th measurements allowed to identify more accurately the genotypes more divergent (Table 4).Based on this results it may be inferred that the fruit yield in soursop has a cyclic behavior during the years.Therefore, it is needed that genotypes should have their fruit yield behavior established for studying the genetic diversity among soursop genotypes, because it is a quantitative trait and consequently its heritability is low.

Figure 2 .
Figure 2. Clustering of the 71 soursop genotypes using the first two principal components (PCA 1 and PCA 2).Groups were established based on Tocher's optimization (each symbol represent one group based on Tocher's optimization -Table3).The experiment was laid out from 2000 to 2016.

Table 1 .
Analysis of variance for fruit yield evaluated in 71 soursop genotypes in 16 measurements.

Table 2 .
Scott-Knott clustering for fruit yield (kg/plant) evaluated in 71 soursop genotypes among 16 measurements.The experiment was laid out from 2000 to 2016.

Table 3 .
Tocher's optimization procedure for 71 soursop genotypes based on 16 measurements for fruit yield.

Table 4 .
Contribution (S.j.) of the 16 fruit yield measurements for genetic diversity evaluated among 71 soursop genotypes based on Singh (1971) criteria.The experiment was laid out from 2000 to 2016.