Genetic diversity analysis and genotpe x environment interaction in Ethiopian mustard (Brassica carinata A. Braun)

[摘要] English: Ethiopian or Abyssinian mustard (Brassica carinata A. Braun) is one of the mostimportant oil crops in Ethiopia. Apart from its phylogenetic relationships with othermembers of the genus Brassica, little information is available for this species. Thisstudy explored the genetic diversity and genotype x environment interaction for itspossible use in genetic conservation and breeding. Geographical patterns ofmorphological variations were studied on 258 accessions of B. carinata randomlysampled from different parts of the country using Shannon Weaver diversity index( H') to estimate the phenotypic diversity index for six morphological traits. Theresults revealed that the mean diversity index (H') across regions of the country was0.72±0.06. Polymorphism was high for some of the traits. The mean diversity index(H') for the country across altitudinal classes was H' = 0.71 ± 0.07. Highest meandiversity index was recorded from areas ranging from 2600-280 masl. The centres ofdiversity appears to be Shewa, Bale and Arsi provinces with due consideration ofaltitudes. Further exploration and in situ conservation efforts should be undertaken inthese regions.Brassica carinata possesses a number of agronomic advantages over other oil seedcrops with similar ecological adaptation. However, the oil is considered low quality.Although efforts have been made to improve its quality, much has to be done to usenatural variations that might exist within the species for fatty acid composition. Theobjectives of this study were to determine the oil content and fatty acid composition todetermine the range of genetic diversity for these traits, and to select genotypes thatcan be used in crosses to improve these traits. Ninety-eight Ethiopian growngenotypes were analyzed by CGC to study the variability of fatty acid composition.The analysis revealed a wide variation in fatty acid composition; twenty-six fattyacids were identified. The predominant fatty acids were erucic (6.91-35.05%), linoleic(17.21-28.53%),α -linolenic (10.50-22.52%), and oleic (8.91-24.64%), followed bygadoleic (1.22-10.24%) and palmitic (3.38-16.42%). Palmitic acid had significantpositive correlations with stearic acid. Erucic acid showed significant and negativelinear correlations with palmitic, stearic, vaccinic, oleic, linoleic, α-linolenic acids andpositive correlation with eicosenoic acid. The high oil yielding genotypes wererelatively richer in erucic acid content. Oil content ranged from 25 to 48%.Genetic diversity analysis within B. carinata has not been examined with molecularmarkers. The objective of this study was to investigate the genetic relationshipsamong 39 B. carinata genotypes by the AFLP markers. This study explored theamount of genetic variation for possible use in breeding. Thirty-nine genotypes of B.carinata were analysed using six AFLP primer combinations. A total of 189polymorphic markers were scored, with an average of 32 markers per primercombination. GDE ranged from 0.346 to 0.639, with a mean of 0.504 ± 0.002.Polymorphism rates ranged from 50 to 80%. UPGMA cluster analysis revealed sevengenetically distinct groups of genotypes.The GE interaction and the stability of 14 B. carinata genotypes for seed yield wereevaluated at four rain-fed locations, over three years in north-western Ethiopia. Theobjectives were to estimate the component of variance associated with the first andsecond order interactions and to determine their effects and to measure the genotypes'stability using different stability statistics; and to compare the stability statistics.The analysis of variance test revealed that the main effects due to years, locations, andthe first order interactions (year x location) as well as second order interactions(genotype x year x location) were highly significant. The genotype x locationinteraction was not significant. This suggests the necessity for increased emphasis onmultiple years of testing than more locations. The estimates of variance componentsinvolving first order interaction of genotype x year, genotype x location and the errorvariance were very small. The variance components of the main effects of thegenotype and second order interaction of genotype x location x year were very high.positive correlation with eicosenoic acid. The high oil yielding genotypes wererelatively richer in erucic acid content. Oil content ranged from 25 to 48%.Genetic diversity analysis within B. carinata has not been examined with molecularmarkers. The objective of this study was to investigate the genetic relationshipsamong 39 B. carinata genotypes by the AFLP markers. This study explored theamount of genetic variation for possible use in breeding. Thirty-nine genotypes of B.carinata were analysed using six AFLP primer combinations. A total of 189polymorphic markers were scored, with an average of 32 markers per primercombination. GDE ranged from 0.346 to 0.639, with a mean of 0.504 ± 0.002.Polymorphism rates ranged from 50 to 80%. UPGMA cluster analysis revealed sevengenetically distinct groups of genotypes.The GE interaction and the stability of 14 B. carinata genotypes for seed yield wereevaluated at four rain-fed locations, over three years in north-western Ethiopia. Theobjectives were to estimate the component of variance associated with the first andsecond order interactions and to determine their effects and to measure the genotypes'stability using different stability statistics; and to compare the stability statistics.Mean yield over all locations and years ranged from 1304 to 1541 kg/ha by PGRCIE210406 and PGRCIE 21163 respectively. The regression of yield on theenvironmental index resulted in a regression coefficient value ranging from 0.7488-1.2288. The deviation from regression was non significant for all genotypes. The firsttwo interaction principal component axes of the additive main effects andmultiplicative interaction (AMMI) model together accounted between 35.7% and54.3% of the total genotype x environment interaction of sum of squares for seedyield. The stability analyses identified PGRC/E 20017 and PGRC/E 20112 as morestable genotypes, while PGRC/E 21169 and (4DxZEM-1) x (ZEM-1-AD/88) werespecifically adapted to some environments.Spearman's coefficient of rank correlation was performed for each of the possible pairwise comparisons of different stability statistics. Significant rank correlations wereobserved among some of the stability statistics. These significant rank correlationsindicated their effectiveness in detecting stable genotypes over a range ofenvironments. The hierarchical clustering of AMMI adjusted mean yield usingUPGMA clustering method grouped the genotypes into five genetically distinctclusters and the environments into three heterogeneous clusters.

[发布机构] University of the Free State