Fig. 3: Evaluation of LC-MS1 data processing and quantification.
From: MetaboAnalystR 4.0: a unified LC-MS workflow for global metabolomics
a Design of the metabolomics experiment. b Statistics of LC-MS1 features across different modes. Generic features, MS features detected across all blood samples; Unique features, features detected only in one type of blood samples. c PCA score plot of the blood dataset (C18, ESI+). d Heatmap of the complete metabolic profiles (MS1 level, C18, ESI+). Unique MS1 features for a specific blood sample type were highlighted with rectangles. Blue, unique features for serum compared to plasma; Green, unique features for plasma compared to serum. Ruby, unique features for whole blood compared to plasma and serum. e Design of serial dilutions. Urine and serum are mixed according to the ratio labeled at the x-axis. f Correlation analysis of MS features from serial dilutions. MetaboAnalystR reported high average correlation coefficients (>0.85) across different modes, compared to other tools (Supplementary Fig. 7).
