Bioinformatical evaluation of modified nucleosides as biomedical markers in diagnosis of breast cancer

被引:47
作者
Bullinger, Dino [1 ]
Froehlich, Holger [2 ]
Klaus, Fabian [3 ]
Neubauer, Hens [4 ]
Frickenschmidt, Antje [1 ]
Henneges, Carsten [2 ]
Zell, Andreas [2 ]
Laufer, Stefan [5 ]
Gleiter, Christoph H. [1 ]
Liebich, Hartmut [3 ]
Kammerer, Bernd [1 ]
机构
[1] Univ Tubingen Hosp, Dept Pharmacol & Toxicol, Div Clin Pharmacol, D-72076 Tubingen, Germany
[2] Ctr Bioinformat Tubingen ZBIT, D-72076 Tubingen, Germany
[3] Univ Tubingen Hosp, Med Clin, D-72076 Tubingen, Germany
[4] Univ Tubingen Hosp, Dept Obstet & Gynecol, D-72076 Tubingen, Germany
[5] Univ Tubingen, Inst Pharm, D-72076 Tubingen, Germany
关键词
metabolomics; high performance liquid; chromatography with ultraviolett detection (HPLC-UV); nucleosides; breast cancer; Support Vector Machine (SVM); k-nearest-neighbor classifier (k-NN);
D O I
10.1016/j.aca.2008.04.048
中图分类号
O65 [分析化学];
学科分类号
070302 ; 081704 ;
摘要
It is known that patients suffering from cancer diseases excrete increased amounts of modified nucleosides with their urine. Especially methylated nucleosides have been proposed to be potential tumor markers for early diagnosis of cancer. For determination of nucleosides in randomly collected urine samples, the nucleosides were extracted using affinity chromatography and then analyzed via reversed phase high-performance liquid chromatography (HPLC) with UV-detection. Eleven nucleosides were quantified in urine samples from 51 breast cancer patients and 65 healthy women. The measured concentrations were used to train a Support Vector Machine (SVM) and a k-nearest-neighbor classifier (k-NN) to discriminate between healthy control subjects and patients suffering from breast cancer. Evaluations of the learned models by computing the leave-one-out error and the prediction error on an independent test set of 29 subjects (15 healthy, 14 breast cancer patients) showed that by using the eleven nucleosides, the occurrence of breast cancer could be forecasted with 86% specificity and 94% sensitivity when using an SVM and 86% for both specificity and sensitivity with the k-NN model. (C) 2008 Elsevier B.V. All rights reserved.
引用
收藏
页码:29 / 34
页数:6
相关论文
共 32 条
[1]  
[Anonymous], MODIFICATION EDITING
[2]   SERUM CREATININE DETERMINATION WITHOUT PROTEIN PRECIPITATION [J].
BARTELS, H ;
BOHMER, M ;
HEIERLI, C .
CLINICA CHIMICA ACTA, 1972, 37 (NMAR) :193-&
[3]   USING MUTUAL INFORMATION FOR SELECTING FEATURES IN SUPERVISED NEURAL-NET LEARNING [J].
BATTITI, R .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 1994, 5 (04) :537-550
[4]   TRANSFER-RNA MODIFICATION [J].
BJORK, GR ;
ERICSON, JU ;
GUSTAFSSON, CED ;
HAGERVALL, TG ;
JONSSON, YH ;
WIKSTROM, PM .
ANNUAL REVIEW OF BIOCHEMISTRY, 1987, 56 :263-287
[5]  
Bullinger D, 2005, LC GC EUR, P16
[6]  
CORTES C, 1995, MACH LEARN, V20, P273, DOI 10.1023/A:1022627411411
[7]  
Duda RO, 2006, PATTERN CLASSIFICATI
[8]  
Dudley E, 2000, RAPID COMMUN MASS SP, V14, P1200, DOI 10.1002/1097-0231(20000730)14:14<1200::AID-RCM10>3.0.CO
[9]  
2-I
[10]  
Fan XH, 2005, P ANN INT IEEE EMBS, P6081