Analysis of quantitative variation in conservation and utilization of Ornamental genetic resources.pptx

Conservation

1. UNIVERSITY OF HORTICULTURAL SCIENCES, BAGALKOT (Karnataka-India) HORTICULTURE RESEARCH AND EXTENTION CENTRE HOGALAGERE Dr. R.K. RAMACHANDRA, M.Sc. (Agri.), Ph.D. Assistant Professor of Genetics and Plant Breeding [phone] (Mobile), Email ids , [email protected] , [email protected] Analysis of quantitative variation in conservation and utilization of Ornamental genetic resources

2. Biometrics Assessment of genetic variability existing in the population : various biometrical tools are used for this viz range, mean , variance, standard deviation, co efficient of variation, standard error, D 2 statistics and metroglyph analysis Selection of elite genotype from mixed population : Here various tools are used namely correlation coefficient , path analysis, and discriminate function analysis Choice of the parents for hybridization : Various tools such as Diallel ( Haymans graphical approach and griffings numerical approach) , Partial diallel ( Kemphthorne ), line X tester or modified top cross , generation mean analysis ( Haymans ) , triallel , quadri allel ( Rawling and cockerham ), biparntal cross(Comstock and Robinson) Determining the varietal adaptation : stability analysis

3. Biometrical analysis Mean, Variance, Standard deviation: Standard Error (SE) Coefficient of Variation (CV): Correlation coefficient (r Regression coefficient (b): Covariance: Heritability Genetic Advance PCV, GCV Percentage of Homozygosity , estimation of number of crosses with ‘n’ inbreds / varietiesgosity :

4. Generation of data- Experimental design Formal layout plan for allocation of treatments to different plots with the following objectives Objectives obtain unbiased estimates of treatment means provide an estimate of experimental error test the significance of difference among treatment means reduce the magnitude of experimental error Experimental error Variation among experimental units treated alike Sources of experimental error Plant to plant variation Plot to plot variation Seasonal effects Measurement errors Principles of experimental designs Replication, Randomization, Blocking,Local control

5. Choice of experimental design Designs differ from each other w.r.t . the amount of restriction they impose on random allocation of treatments to different plots Always choose the simplest experimental design Stage of breeding programme Number of treatments to be evaluated Fertility gradient Number of factors to be evaluated Advantages of CRD Simplest in adoption and analysis Flexibility - unequal number of replications can be easily handled Maximum error df Disadvantages Low precision if plots are not uniform Uses Expt. site is relatively uniform or recognizable basis of grouping absent Large number of plots may be lost Limited number of plots

6. Layout and ANOVA(CRD) A B C B A A D C B C D A D C D B Source df MS Treatments t-1 MST Error t(r-1) MSE Total tr-1

7. Randomized Complete Block Design Most commonly used design Expt. units are classified into blocks Block should be square in absence of gradient Block should be long & narrow and perpendicular to gradient Each block is complete – carries all treatments Each treatment appears once in each block Blocks = replications Each block is handled as one unit Treatments assigned at random within each block Separate randomization for each block Advantages of RBD : Controls one source of variation, Smaller magnitude of error mean squares, Increased precision, Simple in adoption and analysis Disadvantages: Reduced error df Uses : Expt. site has unidirectional gradient, Limited number of treatments (4-20)

8. Layout and ANOVA(RBD) A B C D A C D B B D C A D C A B Source df MS Blocks r-1 MSB Treatments t-1 MST Error (t-1)(r-1) MSE Total tr-1

9. P = G + E + GE Contribution of G & E is interdependent Same genotype exhibits different phenotype in different envts . Effect of a particular environment is not exactly same on two or more genotypes GE is present as a rule rather than an exception “It is not too much to say that GE, not the genotype, is the cultivar” Jensen (1988) Consequences of GE Performance of a genotype in any envt is difficult to predict Individuals selected in one envt may perform very poor in another envt True genotypic potential can be estimated after multienvt testing Influences breeding objectives and strategies Type of variety to be developed Type of envt for selection – stress vs non-stress Testing sites

10. Augmented designs Under the following two situations, RBD or any such design becomes inadequate or inefficient. a) When seeds of test genotypes are small in quantity so that replications are not possible. For instance, some newly introduced stocks, crossed seeds . b) bWhen large number of germplasm collections are to be evaluated, limited facility of uniform land becomes a bottleneck. Such a huge collection (in 100’s or 1000’s) apparently cannot be accommodated in RBD as soil heterogeneity becomes unmanageable. In absence of error term (error variance) in the ANOVA, without replication, test of significance cannot be applied. :

11. Augmented Design I When number of seeds is a limitation, the augmented completely randomized design I is most suited. In this design, the whole experimental area is divided into N plots. Where, N is number of test genotypes (v) + number of checks (c) which are standard varieties or hybrids of known performance, repeated b times i.e., N = v + bc ; Blocks are not marked at all. A representative field design with v = 8 (v 1 to v 8 ) c = 4 (c 1 to c 4 ), b = 3, e =12 and N = 20 in a linseed trial is depicted below : C3 V6 C4 C3 V1 C4 V7 C4 V4 C2 C1 V3 C2 C1 C3 V5 C1 V8 C2 V2

12. Field plan and randomization: This can be illustrated using the example given in augmented design I, i.e., v = 8 (v 1 to v 8 ), c = 4 (c 1 to c 4 ), b = 3, c = 12 and N = 20. Now divide the whole experimental field into (b = 3) distinct blocks. Block I, consisting of 7 plots (c 1 to c 4 + v 1 , v 4 and v 8 ), Block II, comprising 6 plots (c 1 to c 4 + v 2 and v 3 ), Block III, comprising 7 plots (c 1 to c 4 + v 5 , v 6 and v 7 ) C2 C4 V8 C1 V1 C3 V4 V3 C3 C2 V6 C1 C4 V7 C3 C2 V5 C1 V2 C4

13. CALCULATION PART range, mean , variance, standard deviation, co efficient of variation correction factor, total sum of square, treatment sum of square. replication sum of square , make anova table , calculate critical difference standard error, f value, environmental variance, genotypic variance, phenotypic variance, heritability, phenotypic co efficient of correlation ,genotypic co efficient of correlation, genetic advance under selection from following data  

14. Variety / Treatment R I (x) R II (x) R III (x) Total (T j ) Mean (g j ) A 40 45 40 125 41.6 B 50 50 50 150 50.0 C 55 55 50 160 53.3 D 60 65 65 190 63.3 E 40 40 40 120 40.0 F 50 55 55 160 53.3 G 65 60 70 195 65.0 Total (RT j ) 360 370 370 1100 (GT) 52.3 -(X)

15. Stage-1 1) calculation of Correction Factor = (GT) 2 /N where GT is grand total and N is number of observation or R xT 2) calculation of total sum of square(TSS) = (a+ b+c ……….z) 2 - CF 3) treatment sum of square( TrSS ) = ( p+q+r )/ no.replications – cf 4) replication sum of square(RSS) = ( x+y+z )/ no. treatments- cf 5) error sum of square = TSS- TrSS +RSS let us all these four from (2-5) as A B C D respectively

16. SV DF TSS MSS Cal F Table F Replication (R) r-1 Put C value here ( I) I/r-1(L) L/N Treatment/genotype(T) t-1 Put B value here (J) J/t-1(M) M/N Error r-1xt-1 Put D here(K) D/r-1x t-1 (N) -

17. Stage 3   1) Error variance ( Ve ) = from table N value 2) Genotypic variance(Vg) = from table M-N/R 3) Phenotypic variance( Vp ) = combination pf these above two Vg+ve 4) heritability = Vg/ Vp Stage -4   PCV (Phenotypic coefficient of variation)= standard deviation of Vp / mean X 100 GCV(Genotypic coefficient of variation )= standard deviation of VG/mean X 100 GAS Genetic advance under selection = K p h 2 GAM Genetic advance as % of mean = GAS/ mean X 100

18. Both heritability and genetic advance High heritability accompanied with high genetic advance: it indicates that most likely the heritability is due to additive gene effects and selection may be effective. High heritability accompanied with low genetic advance: It is indicative of non-additive gene action. The high heritability is being exhibited due to favourable influence of environment rather than genotype and selection for such traits may not be rewarding. Low heritability accompanied with high genetic advance: It reveals that the character is governed by additive gene effects. The low heritability is being exhibited due to high environmental effects. Selection may be effective in such cases. Low heritability accompanied with low genetic advance: It includes that the character is highly influenced by environmental effects and selection would be ineffective.

19. Combining ability : Ability of a strain to produce superior progeny when crossed with other strains. General combining ability : Average performance of a strain in a series of cross combinations. The GCA is estimated from the performance of F1s from the crosses. The tester will have a broad genetic base. A2 B2 C2 MEAN GCA A1 50 40 60 50 -20 B1 60 70 50 60 -10 C1 100 110 90 100 +30 MEAN 70 73.3 66.6 70 GCA 3.3 -3.3

20. The GCA effects of A1 = 50-70= -20 The GCA effects of B1=60-70= -10 The GCA effects of C1= 100-70 = +30 Similarly the the GCA effects of A2, B2, and C2 is 0, 3.3, -3.3 respectively Specific combining ability : Deviation in performance of a cross combination from that predicted on the basis of general combining ability of the parents involved in the cross. The testing will be on inbred. SCA effects of A1 x A2 ie GCA of A1=50, A2=0n A1A2= 50-(-20)-0-70 ie = 0 Similarly the C1B2= 110, GCA of C1=30, GCAB2= 3.3 , Therefore the SCA of C1B2= 110-(+30)-3.3-70 110-30-73.3 110-103.3 6.6

21. Chi-square test A dihybrid cross between two pea plants heterozygous for shape and color ( RrYy x RrYy ) produced the following results: Round yellow 315 Round green 108 Wrinkled yellow 101 Wrinkled green 32 Are these results consistent with Mendel’s second law? Observed 315 108 101 32 Expected 312.75 104.25 104.25 34.75 Deviation -2.25 +3.75 -3.25 -2.75 [ 0 – E ] 2 5.06 14.06 10.56 7.56 ( 0 – E ) 2 /E 0.016 0.13 0.10 0.21

22. Chi-square test Generation O.mean E.mean O-E (O-E) 2 .Wt P1 69.44 68.97 0.47 0.3698 P2 59.04 59.95 -0.91 1.2602 F1 83.44 83.23 0.21 0.0851 F2 74.36 73.85 0.51 0.2582 B1 76.02 76.10 -0.08 0.0079 B2 71.28 71.59 -0.31 0.1058 Chi square = 2.0870

23. Thank you

24. Diversity analysis Meteroglyph analysis D2 analysis Principle component analysis NTSys, Darwin, Phylip, etc.

25. METEROGLYPH ANALYSIS Semigraphic method for assessing the pattern of morphological variation in a large number of germplasm lines taken at a time. First order statistics more reliable and robust Very simple From both replicated as well as non replicated data. Pattern of variability is depicted by glyph on the graph.

26. Example Five characters of cotton from 35 genotypes - seed cotton yield per plant, boll weight, number of bolls per plant, ginning outturn and mean length.

27. D 2 analysis This is a numerical approach which is used for measuring genetic divergence in the germplasm collections. Estimates of D2 statistics are based on second order statistics and therefore, have lesser precision than metroglyph analysis. Analysis is more difficult than metroglyph analysis. Analysis is possible from replicated data only. Genetic diversity is depicted by cluster diagram.

28. In the selection of parents on the basis of D2 values, three important points should be taken into consideration. choice of clusters with maximum genetic distance, selection of one or two genotypes from such clusters. relative contribution of each character to the total divergence,

29. NTSYSpc – Programme for multivariate analysis Developed for use in biology in the context of the field of numerical taxonomy NTSYS— N umerical T axonomy SYS tem NTSYSpc is a system of programs that is used to find and display structure in multivariate data. widely used in morphometrics , ecology and in many other disciplines in the natural sciences, engineering, and the humanities used to compute various measures of similarity or dissimilarity between all pairs of objects. DARWIN software for diversity analysis DARWin stands for Dissimilarity Analysis and Representation for Windows. Current version is 5.

30. CORRELATION IN PLANT BREEDING Correlation coefficient is a statistical measure find out the degree (Strength) and direction of relationship between two or more variables. It is independent of the unit of measurement. Its value lies between – 1 and 1. It measures the degree and direction of association between two or more variable.

31. SIMPLE CORRELATION The association between any two variables is termed as simple correlation or total correlation or zero order correlation coefficient. Phenotypic correlation It includes both genotypic and environmental effects and, therefore, differs under different environmental conditions. Genotypic correlation The inherent or heritable association between two variables is known as genotypic correlation. This type correlation may be either due to pleiotropic action of genes or due to linkage or more likely both. Environmental Correlation It is due to error variance It is not heritable.

32. Computation of Correlation All the three of correlation,partial be estimated from both unreplicated and replicated data. But phenotypic, genotypic and environmental correlations can be estimated from replicated data only. All possible combinations of correlations n (n -1) /2, where n is the number of variables or characters

33. Test of significance The calculated r can be tested for its significance by comparing it with the table value of r available in standard books at n-2 degrees of freedom. In the absence of r table values, the test of significance is accompanied by t test as under: t= r/ SE where, SE= √ [ (1-r 2 )/(n-2) ]

34. Interpretation of simple correlations If the value of r is significant, it indicates the existence of association between two characters. If the value of r bears negative (-) sign it means that increase in one character will lead to decrease in second and vice versa. Similarly, if it bears the positive(+) sign it means that increase in one variable will cause increase in the other and vice versa.

35. Rank correlation When the distribution of variables is not normal, the degree of relationship between the variables can be determined using Rank correlation . It is the study of relationships between different rankings on the same set of items. A rank correlation coefficient measures the correspondence between two rankings and assesses its significance. Spearman rank correlation test is used when the Pearson correation gives misleading results.

36. Path analysis It is a partial regression It requires dependant and independent variables. It partitions the correlation coefficient into direct and indirect effect of each independent variable on the dependant variable. The residual effect need to be near zero. Maximum value is 1. Nearer to zero indicates most appropriateness of the characters used in path analysis. >1 = very high ; 0.3 to 1 = high; 0.2 to 0.3 = medium 0.1 to 0.2 = low; <0.1 = negligible

37. Mating Design Top cross Poly cross Line X tester crosses Diallel Generation mean analysis Triple Test Cross Three Way cross North Carolina Design, etc.

38. Top cross One parent is crossed with many parents. Mostly used to assess the combining ability of the lines or inbreds . Good to use in early generation testing of progenies. General combining ability alone can be obtained. Mid parental, better parent and standard heterosis is possible.

39. Poly cross This method is similar to top cross but it assumes a random male parentage. It has wider genetic back ground of pollen source. Eg . Open pollination of large number some parental lines. Pollen source is random. Evaluation of flower harvested from each line for their combining ability

40. Line X Tester analysis It is the extension of Top cross method. In case of top cross only one tester is used, while in case of line x tester cross several tester are used. The common parents is referred to as the tester and the hybrids produced are known as test crosses or top crosses. The tester may be male or female and no restriction.

41. Line X Tester analysis Female Parents Male parents m1 m2 m3 m4 m5 f1 X X X X X f2 X X X X X f3 X X X X X f4 X X X X X f5 X X X X X f6 X X X X X f7 X X X X X f8 X X X X X f9 X X X X X f10 X X X X X CASE STUDY

42. TYPES OF DIALLEL CROS S Full Diallel Full diallel with parents (F1 s, Reciprocals and Parents ) includes both way crosses and parents The total number of entries to be evaluated is equal to p 2 , where p is the number of parents included in a diallel used when parents do not have self incompatibility or male sterility The inclusion of parents permits the estimation of heterosis.

43. Full Diallel without Parents (F 1 s and Reciprocal) includes all possible single crosses made among ‘p’ parents in both the direction i.e., direct and reciprocal crosses The total number of single crosses is equal to p(p-1) used when the presences of self incompatibility or male sterility prevents the inclusion of parents in the experiment. does not permit estimation of heterosis.

44. HALF DIALLEL Half diallel with parents (F1s and parents) includes one way crosses and parents The total number of entries to be evaluated is equal to p(p+1)/2 , where p is the number of parents involved in the mating This can be used when parents do not have self incompatibility or male sterility The inclusion of parents in the experiment permits the estimates of heterosis.

45. Half Diallel without Parents (F1s Only) includes all possible single crosses made in the one direction only. The number of single crosses required is equal to p(p-1)/2 The design can be used when the presence of self incompatibility or male sterility prevents the inclusion of parents in the experiment Thus estimate of heterosis is not possible in this method .

46. Plan of Crossing for a Diallel Design Parents 1 2 3 4 5 6 7 8 9 10 1 * X X X X X X X X X 2 + * X X X X X X X X 3 + + * X X X X X X X 4 + + + * X X X X X X 5 + + + + * X X X X X 6 + + + + + * X X X X 7 + + + + + + * X X X 8 + + + + + + + * X X 9 + + + + + + + + * X 10 + + + + + + + + + * Where, X, +, and * = direct crosses, reciprocals, and parents (selfing), respectively.

47. Partition of variation and Degree of Freedom in Four Methods of Diallel Analysis Sources of Variation Method 1 F 1 s + P + R Method 2 F 1 s + Parents Method 3 F 1 s + R Method 4 F 1 s only Replications r-1 r 1 r Treatments t-1 t-1 t-1 t-1 gca p-1 p-1 p-1 p-1 sca c c c-p c-p Reciprocals c - c - Error (r-1) (t-1) (r-1) (t-1) (r-1) (t-1) (r-1) (t-1) CASE STUDY

48. Helps to assess the gene action. Mean of various generations like P1, P2, F1, F2, B1, B2 and F3 will be used to assess the gene action. If all the scales i.e., A=B=C=D=0, then additive and dominance gene action present in that cross for the character. Any deviation is considered as the presence of epistasis . SEA =  VA  SEB =  VB  SEC =  VC SED =  VD   t=A/SEA A = 2B1-P1-F1 B = 2B 2 – P 2 -F 1 C = 4F 2 – 2F 1 – P 1 – P 2 D = 4F 3 – 2F 2 –P 1 -P 2 GENERATION MEAN ANALYSIS

49. Stability analysis reflects the suitability of a variety for general cultivation over a wide range of environments. Performance of a genotype mainly depends on environmental interaction. Estimation of phenotypic stability involves regression analysis. assess the response of various genotypes under changing environment conditions. Genotype-environment interactions gives an idea of the buffering capacity of the genotypes. Low genotype environmental interactions indicates consistent performance of a population over variable environments.

50. Models of stability analysis A. Conventional models Stability factor model (Lewis 1954) Ecovalence model ( Wricke 1964) Stability variance model ( Shukla 1972) Lin and Binns model (198)  B. Regression coefficient model Finlay and Wilkinson model (1963) Eberhart and Russell model (1966) Perkins and Jinks model (1968) Freeman and Perkins model (1971) Genotypic stability model (Tai 1971) C . Principle component analysis Perkins (1972); Freeman and Dowkar (1975); Seif et al (1979) Additive main effect and multiplicative interaction effect (AMMI) ( Gauch 1988); Freeman (1990) and Zobel (1990) D. Cluster analysis Grouping by cluster analysis ( Westcot 1987) Webber and Wricke (1990) Francis and Kannenberg (1975)  

51. Finlay and Wilkinson (1963) The first systematic approach to the analysis of phenotypic stability of cultivars or genotype was made by Finlay and Wilkinson in 1963. They used two parameters namely (1) mean performance over the environments. (2) regression of performance in different environments over the respective environmental mean. The following inferences can be drawn from the analysis: a) b=0 indicates stability b) b>1 or b<1 indicates that the genotype is not stable where ‘b’ is regression coefficient

52. Eberhart and Russell (1966) A stable variety as one with a regression coefficient of unity (b=1) and a minimum deviation from the regression line (s 2 d=0) The following inferences can be drawn from the analysis:   1) If the genotype has non significant s 2 d, then that genotype can be classified as stable genotype. 2) The regression coefficient is used to assess the response to the environments .

53. ANOVA for Stability - Eberhart and Russell Source of variation Df Genotypes g-1 Environemnts + GXE G( e-1) Environment (linear) 1 Gx E (linear) g-1 Pooled deivation G(e-2) Genotype 1 e-2 Genotype 2 e-2 …. ….. Genotype n e-2 Pooled error Ge (r-1)

54. Inference on regression coefficients: Result on regression coefficient Inference b=1 and NS Average stability Suitable to all environment b>1 and Significant Below average stability Suitable for specific environment b<1 and Significant Greater resistance to environmental changes. Above average stability Suitable to poor environment. b=0 and NS No response to the environment Non responsive genotype.

55. Additive Main effect and Multiplicative Interaction (AMMI) model The AMMI model first calculates genotypes and environment additive (main) effects using analysis of variance (ANOVA) and analyses the residual from this model (namely the interaction) using principal components analysis (PCA).

56. ANOVA computes a genotype deviation (difference from the grand mean) and an environment deviation PCA computes genotype score and environment score PCA to decompose the interaction into PCA axes 1 to N, and a residual ρ ge remains if not all axes are used. These PCA scores are termed as interaction PCA scores or IPCA scores.

57. SOFTWARE Genres Agres TNAUSTAT SPARE (IASRI) – SPARE II (window based) INDOSTAT - Windostat SPSS Cropstat (IRRI) NTYSys-pc Darwin Mapmaker QTL cartographer Imas

58. Temperature trend 21 st century T hanks…