**3. Identification and characterization of secondary metabolites**

The identification and characterization of secondary metabolites are important. Metabolomic often requires abroad array of instrumentation such as ELSD for detecting lipids, coulometric array detectors for detecting redox compounds, and fluorescent spectrometer for detecting aromatic compounds, whereas other omics techniques such as genomics, transcriptomics, or proteomics are often conducted by a single instrument.

In microbial secondary metabolites investigation, the experiments are mainly conducted in two different approaches, targeted or untargeted metabolites identification [34]. As its name, targeted metabolites experiment aims to identify a specific group of SMs that are already known. Whereas, the untargeted secondary metabolites experiment aims to identify the large scale of SMs produced by microorganisms including novel and known metabolites [35].

Nowadays, two general technologies have been utilized as primary tools in metabolomic, mass spectrometry (MS), and nuclear magnetic resonance (NMR) [4, 36, 37].

These high-throughput tools provide broad coverage of many classes of secondary metabolites, including amino acids, lipids, sugars, organic acids, and others.

In fact, nuclear magnetic resonance (NMR) and mass spectrometry (MS) has been used to identify both targeted and untargeted secondary metabolites [38]. They are often complementary to each other. Mass spectrometry (MS) provides information of molecules whereas, nuclear magnetic resonance (NMR) is utilized to differentiate between structural isomers [39]. In fact, MS is more sensitive than NMR and able to detect the large scale of metabolites, while NMR is highly quantitative and reproducible and require larger sample amount for analysis than MS [40, 41].

#### **4. Data analysis**

In fact, the major challenges in metabolomic experiments are the huge amount of information obtained from either NMR spectroscopy or MS [7, 37]. The extraction of the significant information generated by NMR and MS is crucial by using computer software in order to organize the vast amount of data [40, 42].

Because studying individual metabolites is impractical for visualizing changes between groups of metabolites, univariate statistical approaches can be utilized to understand the results. Principal component analysis (PCA) is one of the most extensively used statistical approaches [39, 43, 44]. The data can be simplified using principle component analysis. CA without losing its core feature. In fact, principal component analysis PCA provides information on multivariate differences among secondary metabolites while, different univariant statistical tests such as non-parametric Wilcoxon signed-rank test, Kruskal–Wallis test, and the parametric.

Student's t-test and ANOVA can be utilized to analyze isolated metabolites [45].

Nowadays, most metabolites can be identified, due to the development of many bioinformatics software. There are two types of metabolites identification that are applied including 1) definitive identification and 2) putative identification [7]. Many different metabolomics databases are available online some of them are used for NMR such as METLIN (http://metlin.scripps.edu), Biological Magnetic Resonance Databank (http://www.bmrb.wisc.edu/metabolomics/), and METLIN (http://metlin.scripps.edu) while the others are used for MS such as Mass Bank (http://www.massbank.jp), http:// csbdb.mpimp-golm.mpg.de/csbdb/gmd/gmd.html"http://csbdb.mpimp-golm.mpg.de/ csbdb/gmd/gmd.html), the Glom Metabolite Database (GMD, NIST (http://www.nist. gov/srd/nist1a.htm), METLI and MMCD (http://mmcd.nmrfam.wisc.edu) [46].

#### **5. Conclusion**

Microorganisms are one of the most significant sources of SMs that play important roles in many aspects of our life including pharmaceutical, biomedical and food
