Multivariate statistical techniques are used extensively in metabolomics studies, ranging from biomarker selection to model building and validation. Two model independent variable selection techniques, principal component analysis and two sample t-tests are discussed in this chapter, as well as classification and regression models and model related variable selection techniques, including partial least squares, logistic regression, support vector machine, and random forest. Model evaluation and validation methods, such as leave-one-out cross-validation, Monte Carlo cross-validation, and receiver operating characteristic analysis, are introduced with an emphasis to avoid over-fitting the data. The advantages and the limitations of the statistical techniques are also discussed in this chapter.

Hepatitis C virus (HCV) infection of the liver is a global health problem and a major risk factor for the development of hepatocellular carcinoma (HCC). Sensitive methods are needed for the improved and earlier detection of HCC, which would provide better therapy options. Metabolic profiling of the high risk population (HCV patients) and those with HCC provides insights into the process of liver carcinogenesis and possible biomarkers for earlier cancer detection. Seventy-three blood metabolites were quantitatively profiled in HCC (n=30) and cirrhotic HCV (n=22) patients using a targeted approach based on liquid chromatography resolved tandem mass spectrometry (LC-MS/MS). Sixteen of the 73 targeted metabolites differed significantly (p < 0.05) and their levels varied up to a factor of 3.3 between HCC and HCV. Four of these 16 metabolites (methionine, 5-hydroxymethyl-2′-deoxyuridine, N2,N2-dimethylguanosine and uric acid) that showed the lowest p-values were used to develop and internally validate a classification model using partial least squares discriminant analysis. The model exhibited high classification accuracy for distinguishing the two groups with sensitivity, specificity and area under the receiver operating characteristic curve of 97%, 95%, and 0.98, respectively. A number of perturbed metabolic pathways including amino acid, purine and nucleotide metabolism were identified based on the 16 biomarker candidates. These results provide a promising methodology to distinguish cirrhotic HCV patients, who are at high risk to develop HCC, from those who have already progressed to HCC. The results also provide insights into the altered metabolism between HCC and HCV.

TMEM120A, also named as TACAN, is a novel membrane protein highly conserved in vertebrates and was recently proposed to be a mechanosensitive channel involved in sensing mechanical pain. Here we present the single particle cryo-EM structure of human TMEM120A which forms a tightly packed dimer with extensive interactions mediate by the N-terminal coiled coil domain (CCD), the C-terminal transmembrane domain (TMD), and the re-entrant loop between the two domains. The TMD of each TMEM120A subunit contains six transmembrane helices (TMs) and has no clear structural feature of a channel protein. Instead, the six TMs form an α-barrel with a deep pocket where a coenzyme A (CoA) molecule is bound. Intriguingly, some structural features of TMEM120A resemble those of elongase for very long-chain fatty acid (ELOVL) despite low sequence homology between them, pointing to the possibility that TMEM120A may function as an enzyme for fatty acid metabolism, rather than a mechanosensitive channel.

We recently applied gas chromatography coupled to time-of-flight mass spectrometry (GC/TOF-MS) and multivariate statistical analysis to measure biological variation of many metabolites due to environment and genotype in forage and grain samples collected from 50 genetically diverse nongenetically modified (non-GM) DuPont Pioneer commercial maize hybrids grown at six North American locations. In the present study, the metabolome coverage was extended using a core subset of these grain and forage samples employing ultra high pressure liquid chromatography (uHPLC) mass spectrometry (LC/MS). A total of 286 and 857 metabolites were detected in grain and forage samples, respectively, using LC/MS. Multivariate statistical analysis was utilized to compare and correlate the metabolite profiles. Environment had a greater effect on the metabolome than genetic background. The results of this study support and extend previously published insights into the environmental and genetic associated perturbations to the metabolome that are not associated with transgenic modification.

Nucleosides are indicators of the whole-body turnover of transfer RNA. Based on the activity of cancer cells these molecules could potentially be used as cancer biomarkers, and several studies have determined that the metabolic levels of nucleosides are significantly altered in cancer patients compared to control groups. Here we report a targeted metabolite investigation of serum nucleosides in esophageal adenocarcinoma specimens. We quantified eight nucleosides using high-performance liquid chromatography/triple quadrupole mass spectrometry (HPLC/TQMS) and determined that the metabolic levels of 1-methyladenosine (p <2.14 × 10(-7)), N(2),N(2)-dimethylguanosine (p <2.78 × 10(-7)), N(2)-methylguanosine (p <2.48 × 10(-6)) and cytidine (p <6.98 × 10(-4)) were significantly elevated while the concentration of uridine (p <3.74 × 10(-3)) was significantly lowered in serum samples from cancer patients compared to those of control group. Our results suggest that nucleosides could potentially serve as useful biomarkers to identify esophageal adenocarcinoma.