Two Classifiers Based on Serum Peptide Pattern for Prediction of HBV-Induced Liver Cirrhosis Using MALDI-TOF MS
| Autor: | Hai-yan Si, Ming Cheng, Wei Song, Cheng-Jin Hu, Yuan Cao, He-Lin Zhang, Kun He, Na Wang, Ailing Li |
|---|---|
| Rok vydání: | 2013 |
| Předmět: |
Liver Cirrhosis
Proteomics China Hepatitis B virus medicine.medical_specialty Pathology Support Vector Machine Cirrhosis Article Subject lcsh:Medicine medicine.disease_cause Mass spectrometry Sensitivity and Specificity Gastroenterology General Biochemistry Genetics and Molecular Biology Text mining Artificial Intelligence Internal medicine medicine Humans General Immunology and Microbiology medicine.diagnostic_test business.industry lcsh:R Computational Biology Reproducibility of Results General Medicine Hepatitis B medicine.disease Matrix-assisted laser desorption/ionization Case-Control Studies Spectrometry Mass Matrix-Assisted Laser Desorption-Ionization Liver biopsy Peptides business Algorithms Biomarkers Research Article |
| Zdroj: | BioMed Research International, Vol 2013 (2013) BioMed Research International |
| ISSN: | 2314-6141 2314-6133 |
| DOI: | 10.1155/2013/814876 |
| Popis: | Chronic infection with hepatitis B virus (HBV) is associated with the majority of cases of liver cirrhosis (LC) in China. Although liver biopsy is the reference method for evaluation of cirrhosis, it is an invasive procedure with inherent risk. The aim of this study is to discover novel noninvasive specific serum biomarkers for the diagnosis of HBV-induced LC. We performed bead fractionation/MALDI-TOF MS analysis on sera from patients with LC. Thirteen feature peaks which had optimal discriminatory performance were obtained by using support-vector-machine-(SVM-) based strategy. Based on the previous results, five supervised machine learning methods were employed to construct classifiers that discriminated proteomic spectra of patients with HBV-induced LC from those of controls. Here, we describe two novel methods for prediction of HBV-induced LC, termed LC-NB and LC-MLP, respectively. We obtained a sensitivity of 90.9%, a specificity of 94.9%, and overall accuracy of 93.8% on an independent test set. Comparisons with the existing methods showed that LC-NB and LC-MLP held better accuracy. Our study suggests that potential serum biomarkers can be determined for discriminating LC and non-LC cohorts by using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. These two classifiers could be used for clinical practice in HBV-induced LC assessment. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|