3.9 Family 5 of analysis : multivariate analysis

Multivariate analysis can answer to multiple objective. On of then can be to study diversity structure and identify parents to cross based on either good complementarity or similarity for some traits (Figure 3.19). It can be completed by analysis of molecular data and genetic distance trees (M3, section 5).

Decision tree with experimental constraints, designs and methods of agronomic analysis carry out in `PPBstats` regarding the objective :Study diversity structure and identify parents to cross based on either good complementarity or similarity for some traits. **D** refers to designs and **M** to methods.

Figure 3.19: Decision tree with experimental constraints, designs and methods of agronomic analysis carry out in PPBstats regarding the objective :Study diversity structure and identify parents to cross based on either good complementarity or similarity for some traits. D refers to designs and M to methods.

3.9.1 Multivariate analysis (M2)

3.9.1.1 Steps with PPBstats

For variance intra analysis, you can follow these steps (Figure 3.2) :

  • Format the data with format_data_PPBstats()
  • Run the analysis with multivariate()
  • Check outputs and results with functions from factoextra9

3.9.1.2 Format the data

data("data_model_GxE")
data_model_GxE = format_data_PPBstats(data_model_GxE, type = "data_agro")
## data has been formated for PPBstats functions.

3.9.1.3 Run the analysis

vec_variables = c("y1", "y2", "y3")
res.pca = mutlivariate(data_model_GxE, vec_variables, PCA)

3.9.1.4 Check outputs and results

Look at the results thanks to the factoextra package:

fviz_eig(res.pca)

fviz_pca_ind(res.pca, label="none", habillage="location", addEllipses=TRUE, ellipse.level=0.95)