Fig. 1
From: Selection of microbial biomarkers with genetic algorithm and principal component analysis
A diagram of proposed method. s1, s2...sn are the 16S rRNA sequences for this study (can be from other sequencing). v1 to vn are the abundance (normalized) of each species detected in each individual. m = number of PCs created by PCA, n = number of individuals included in the sample. PCA is used to produce PC scores for each individual, and GA is used to select the best subset of PCs to distinguish obesity from healthy cases