Dataset Information

Multi-TGDR, a multi-class regularization method, identifies the metabolic profiles of hepatocellular carcinoma and cirrhosis infected with hepatitis B or hepatitis C virus.

ABSTRACT: Over the last decade, metabolomics has evolved into a mainstream enterprise utilized by many laboratories globally. Like other "omics" data, metabolomics data has the characteristics of a smaller sample size compared to the number of features evaluated. Thus the selection of an optimal subset of features with a supervised classifier is imperative. We extended an existing feature selection algorithm, threshold gradient descent regularization (TGDR), to handle multi-class classification of "omics" data, and proposed two such extensions referred to as multi-TGDR. Both multi-TGDR frameworks were used to analyze a metabolomics dataset that compares the metabolic profiles of hepatocellular carcinoma (HCC) infected with hepatitis B (HBV) or C virus (HCV) with that of cirrhosis induced by HBV/HCV infection; the goal was to improve early-stage diagnosis of HCC.We applied two multi-TGDR frameworks to the HCC metabolomics data that determined TGDR thresholds either globally across classes, or locally for each class. Multi-TGDR global model selected 45 metabolites with a 0% misclassification rate (the error rate on the training data) and had a 3.82% 5-fold cross-validation (CV-5) predictive error rate. Multi-TGDR local selected 48 metabolites with a 0% misclassification rate and a 5.34% CV-5 error rate.One important advantage of multi-TGDR local is that it allows inference for determining which feature is related specifically to the class/classes. Thus, we recommend multi-TGDR local be used because it has similar predictive performance and requires the same computing time as multi-TGDR global, but may provide class-specific inference.

SUBMITTER: Tian S

PROVIDER: S-EPMC4234477 | biostudies-literature | 2014 Apr

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Multi-TGDR, a multi-class regularization method, identifies the metabolic profiles of hepatocellular carcinoma and cirrhosis infected with hepatitis B or hepatitis C virus.

Tian Suyan S Chang Howard H HH Wang Chi C Jiang Jing J Wang Xiaomei X Niu Junqi J

BMC bioinformatics 20140404

<h4>Background</h4>Over the last decade, metabolomics has evolved into a mainstream enterprise utilized by many laboratories globally. Like other "omics" data, metabolomics data has the characteristics of a smaller sample size compared to the number of features evaluated. Thus the selection of an optimal subset of features with a supervised classifier is imperative. We extended an existing feature selection algorithm, threshold gradient descent regularization (TGDR), to handle multi-class classi ...[more]

PMID: 24707821

Similar Datasets

Project description:BackgroundHepatitis C virus (HCV) infection is associated with serum lipid abnormalities, which partly normalize following direct-acting antiviral (DAA) therapy. Here, associations of serum triglycerides (TGs) with viral genotype and markers of liver disease severity were evaluated in patients with chronic HCV. METHODS: The study included the serum of 177 patients with chronic HCV. TGs were quantified by flow injection analysis Fourier transform mass spectrometry. Laboratory values and noninvasive scores for liver fibrosis assessment were determined. The nonparametric Kruskal‒Wallis test, one-way ANOVA, multiple linear regression and Student's t test were used as appropriate. P values were adjusted for multiple comparisons.ResultsHCV-infected women had lower serum TGs than men, and thus, a sex-specific analysis was performed. None of the 46 TG species analyzed differed in the serum of female patients with and without liver cirrhosis. In contrast, in the serum of male patients with liver cirrhosis, TGs with 53, 56 and 58 carbon atoms and three to eight double bonds were diminished. These polyunsaturated TGs were also low in males with a high fibrosis-4 score. TGs with 7 or 8 double bonds negatively correlated with the model of end-stage liver disease score in males. In addition, TGs with 49, 51 and 53 carbon atoms were reduced in male patients infected with genotype 3a in comparison to genotype 1a. TGs with 56 carbon atoms were lower in genotype 3a-infected males than in genotype 1b-infected males. TGs did not differ in females by genotype. Genotype 3-related changes disappeared at the end of therapy with DAAs. Overall, the levels of serum TGs did not change during DAA therapy in either sex. Consequently, the serum TGs of males with liver cirrhosis were lower than those of males without cirrhosis at the end of therapy. Such a difference was not apparent in females.ConclusionsThe decline in TGs observed only in male patients with liver cirrhosis and male patients infected with genotype 3 illustrates sex-specific changes in lipid metabolism in chronic HCV.

Dataset Information

Multi-TGDR, a multi-class regularization method, identifies the metabolic profiles of hepatocellular carcinoma and cirrhosis infected with hepatitis B or hepatitis C virus.

Publications

Multi-TGDR, a multi-class regularization method, identifies the metabolic profiles of hepatocellular carcinoma and cirrhosis infected with hepatitis B or hepatitis C virus.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets