Metabolomics Bioinformatic Analysis Service

The high-throughput methods employed in metabolomics generate substantial amounts of metabolic analysis data, necessitating the use of bioinformatics techniques for its processing. Bioinformatics experts at MtoZ Biolabs are capable of performing thorough and comprehensive analyses of metabolomics data. Beginning with raw mass spectrometry data, they conduct peak alignment, retention time correction, and peak area extraction. Metabolite structure identification is performed using accurate mass matching (within 25 ppm) and secondary spectrum matching in the METLIN and HMDB databases. During comparative analysis of two groups, ion peaks with more than 50% missing values in both groups are excluded. The data is subsequently normalized using autoscaling or the UV method. For statistical analysis, MtoZ Biolabs employs MetaboAnalysis and SIMCA-P software to carry out both multivariate and univariate analyses, including PCA, PLS-DA, OPLS-DA, and pathway enrichment analysis. This approach ensures the delivery of comprehensive and precise metabolomics analysis results.

 

Services at MtoZ Biolabs

1. Quality Assessment of Metabolomics Data

2. Principal Component Analysis (PCA)

3. PLS-DA/OPLS-DA 2D Diagram

4. Data Normalization Analysis

5. Univariate Statistical Analysis

6. Cluster Analysis of Differential Metabolites

7. KEGG Differential Metabolite Pathway Analysis

• • PLS-DA/OPLS-DA Two-Dimensional Diagrams Analysis Service

  In contrast to principal component analysis (PCA), both partial least squares discrimination analysis (PLS-DA) and orthogonal PLS-DA (OPLS-DA) are supervised statistical methods for discriminative analysis. These methods develop models that correlate metabolite expression levels with sample categories to facilitate prediction of the category of samples. 

• • KEGG Differential Metabolite Pathway Analysis Service

  MtoZ Biolabs applies MBRole to enrich metabolic pathways using the differential metabolites, with the KEGG database serving as the reference. The analysis includes all metabolites identified within the same species, focusing on pathways with P < 0.05. Figures 1 and 2 illustrate the types of metabolic pathway diagrams available. These figures depict enriched pathways along with the results of related T-tests.

• • Univariate Statistical Analysis Service

  During the comparative analysis of metabolites between two sample groups, commonly employed univariate methods include Fold Change Analysis (FC Analysis), the T-test, and the Volcano Plot, which synthesizes the preceding two methodologies. This analysis visually elucidates the significance of metabolite variations between samples, aiding in the identification of potential biomarkers (typically selected with FC>2.0 and P value<0.05).

• • Principal Component Analysis (PCA) Service

  Principal Component Analysis (Principal Component Analysis, PCA) is the linear recombination of all metabolites originally identified, forming a set of new comprehensive variables, and selecting 2-3 comprehensive variables from them based on the analyzed problem, to reflect as much information of the original variables as possible, thus achieving the purpose of dimension reduction. At the same time, PCA of metabolites can also reflect the variability between and within groups overall.

• • Metabolomics Data Quality Analysis Service

  MtoZ Biolabs employs two methodologies for the analysis and evaluation of QC samples in project experiments: spectral overlay comparison and principal component analysis (PCA). The spectral overlays from the total ion chromatograms obtained by UPLC-QTOF-MS are compared, as illustrated in the figure below. The results show that the intensities and retention times of the chromatographic peaks are closely aligned, indicating minimal variability due to instrumental errors throughout the experimental process.

• • Differential Metabolites Clustering Analysis Service

  To assess the appropriateness of candidate metabolites and to more comprehensively and intuitively display the relationships among samples and the differential expression patterns of metabolites across different samples, hierarchical clustering is utilized based on the expression levels of significantly differential metabolites. This approach aids in accurately identifying marker metabolites and examining changes in associated metabolic processes.

• • Data Normalization Analysis Service

  Data completeness and accuracy is crucial for subsequent statistical and biological results. After verifying the experimental design and the accuracy of the collected data, MtoZ Biolabs conducts an initial integrity check. This involves removing or imputing missing values, excluding outliers, and normalizing data across samples and metabolites to allow for parallel comparisons. Metabolites with more than 50% missing values in the original data are excluded from further analysis.