Data Processing and Multivariate Analysis (PCA)

Profile Analysis software (version 2.3, Bruker Daltonik GmbH, Germany) was used to preprocess the raw UHPLC−QTOF-MS data. Profile Analysis parameters were set as follows: advanced bucket generation with retention time range of 1.0−25.0 min, mass range of 100−800 m/z, without normalization, with background subtraction and time alignment. LC−MS analyses were processed with the Find Molecular Futures (FMF) function to create compounds (molecular features) with signal-to-noise threshold of 3 for peak detection. Generated bucket table consisting of m/z−retention time pairs and respective compound intensity were imported into Metabolists 4.0 (http://www.MetaboAnalyst.ca/, accessed on 31 January 2021) online software to estimate missing values and to filter and normalize data (normalization by median). No transformation was generalized and data matrix was mean-centered and divided by the square root of the standard deviation of each variable (Pareto scaling). Then, the obtained data matrix was introduced into SIMCA-P+16.01 (Umetrics, Umeå, Sweden) software for in review multivariate statistical analysis of principal component analysis (PCA). The PCA score plot was used to present a natural correlation between the observations. To identify differential compounds, the OPLS-DA (Orthogonal Partial Least Squares Discriminant Analysis) model was used to explore differences in depth between the phytochemical profile of Indonesian and Polish Coleus samples. The OPLS-DA model with VIP values (VIP ≥ 1.0) and |p(corr)| ≥ 0.5 was selected as a differential compound. This provides a preliminary overview of features that are potentially significant for the separation of the two groups (plants cultivated in Poland and Indonesia).