Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

< Back to Article

The Extraction of Simple Relationships in Growth Factor-Specific Multiple-Input and Multiple-Output Systems in Cell-Fate Decisions by Backward Elimination PLS Regression

Figure 5

Reduction of the PLS model by backward elimination PLS regression.

(A) MSE of LOOCV as a function of number of the eliminated variables via the backward elimination PLS regression. Coefficient matrix of the full PLS model with 60 input variables (B), the best PLS model with 22 input variables (C) and the simple PLS model with 5 input variables (D). The red and blue colors indicate positive and negative values, respectively. As the number of the variables reduced, the contribution of remained variables relatively increased, and as a result, magnitude of the regression coefficient increased. The scatter plots of the input loadings (E), input scores (F), output loadings (G) and output scores (H) of the first and second principal components of the simple PLS model. The colors correspond to the latent variables (E, G) and stimuli (F, H).

Figure 5