**4. Conclusions**

In conclusion, the optimum freeze-drying conditions for preserving the nutrients considered in this study and with interesting structural properties of the obtained product, as to be perceived as crunchy by the consumers, are low pressure (5 Pa) and high shelf temperature (50 ◦C). These conditions also promote freeze-dried puree with a clear, yellowish and less saturated colour. The fact that a lower degradation of nutrients was observed at higher temperatures may be explained by the grea<sup>t</sup> reduction (75%) of the duration of freeze-drying process at 50 ◦C, and the mild temperatures used. The shorter exposure of nutrients to a minimal presence of oxygen in a high porous matrix is less favourable to oxidation/degradation reactions and contributes to the preservation of nutrients. As regards to the statistical analysis of the data obtained in this study, PLS-R projection may be recommended against ANOVA as an easier tool to detect the most important factors and interactions to be considered for freeze-drying process optimization. ANOVA allows a more precise analysis, though less practical.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2304-8158/9/1/32/s1, Table S1: Values (mean ± SD) of the di fferent physicochemical properties evaluated. All acronyms used are described in the main text. Table S2: Percentage (mean ± SD) of the bioactive compounds preserved in the FDP for each condition evaluated. All acronyms used are described in the main text. Table S3: Values (mean and Tukey'HSD classification) of the di fferent physicochemical properties and bioactive compounds evaluated for: each individual factor (a–c), the interaction between two different factors (d–f) and for the interaction between the three factors studied (g). All acronyms used are described in the main text. Table S4: Variable importance in the PLS-R Projection (VIP). VIP > 1 are considered as the most important variables for the model. All acronyms used are described in the main text.

**Author Contributions:** Conceptualization: M.A.S.-E., M.d.M.C. and N.M.-N.; Data curation: M.A.S.-E. and M.d.M.C.; Formal analysis: M.A.S.-E., C.A. and N.M.-N.; Funding acquisition: M.A.S.-E., M.d.M.C. and N.M.-N.; Investigation: M.A.S.-E., M.d.M.C. and N.M.-N.; Methodology: M.A.S.-E., C.A., M.d.M.C. and N.M.-N.; Project Administration: M.d.M.C. and N.M.-N.; Supervision, N.M.-N.; Writing-original draft: M.A.S.-E., C.A., and N.M.-N.; Writing-review and editing: M.A.S.-E., C.A., T.F., and N.M.-N. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by Ministerio de Economía y Competitividad of Spain [AGL 2017–89251] and by Ministerio de Educación of Spain [FPU14/02633].

**Acknowledgments:** The authors thank the Ministerio de Economía y Competitividad of Spain for the financial support given through the Project AGL 2017–89251 and the Ministerio de Educación of Spain for the FPU gran<sup>t</sup> (FPU14/02633) awarded to Ms. Andrea Silva.

**Conflicts of Interest:** The authors declare no conflict of interest.
