Fig. 3From: Application of machine learning techniques for creating urban microbial fingerprintst-SNE output to represent microbial profiles on two dimensions. Spearman dissimilarities were calculated from a set of 2239 taxonomic features which represent those present in at least 5% of samples with a minimum relative abundance of 0.1% in a single sample. Confidence regions are 70% confidence regions showing surface typeBack to article page