I collected some data on soil microbial PLFAs from three different rice treatments. Along with some microbial PLFAs, I also collected some data on the soil properties like total nitrogen, soil organic carbon, total protein, etc; My sample size was n=15 from each treatment field. I would like to run a PCA analysis to look for correlations between microbial PLFAs and Soil Properties.

Can I use the mean values of my data to run a Principal Component Analysis? I need some references that can support this. I also ran an ANOVA and saw significant differences between treatments.