Ovarian cancer detection from metabolomic liquid chromatography/mass spectrometry data by support vector machines.
Ovarian cancer detection from metabolomic liquid chromatography/mass spectrometry data by support vector machines.
复制标题
DOI：
10.1186/1471-2105-10-259
复制
发表时间：
2009-08-22
影响因子：
3
通讯作者：
Fernández FM
中科院分区：
文献类型：
--
作者：
Guan W;Zhou M;Hampton CY;Benigno BB;Walker LD;Gray A;McDonald JF;Fernández FM
大多数卵巢癌生物标志物的发现工作集中在识别可以提高目前可用的诊断测试的预测能力的蛋白质。我们在这里表明，代谢组学，研究生物系统中的代谢变化，也可以提供与这种疾病相关的特征性小分子指纹。在这项工作中，新的方法来自动分类卵巢癌患者和良性对照血清产生的代谢组学数据进行了研究。评价了支持向量机（SVM）用于液相色谱/飞行时间质谱（LC/TOF MS）代谢组学数据分类的性能，重点是识别潜在代谢诊断生物标志物的组合或“面板”。应用LC/TOF MS对37例卵巢癌患者和35例良性对照者的血清进行了研究。使用最先进的特征选择方法，如递归特征消除和L1范数SVM，选择了在正或/和负离子模式电喷雾（ESI）MS中观察到的具有区分对照和卵巢癌样本的能力的光谱特征的最佳面板。使用三种评价方法（留一交叉验证、12倍交叉验证、52-20分裂验证）来检查基于所选组的SVM模型区分对照血清样品与疾病血清样品的能力。这些特征选择结果的统计意义进行了全面的调查。血清样本测试集的分类准确率超过90%，这表明上述方法有望开发出一种准确可靠的基于代谢组学的方法来检测卵巢癌。
The majority of ovarian cancer biomarker discovery efforts focus on the identification of proteins that can improve the predictive power of presently available diagnostic tests. We here show that metabolomics, the study of metabolic changes in biological systems, can also provide characteristic small molecule fingerprints related to this disease. In this work, new approaches to automatic classification of metabolomic data produced from sera of ovarian cancer patients and benign controls are investigated. The performance of support vector machines (SVM) for the classification of liquid chromatography/time-of-flight mass spectrometry (LC/TOF MS) metabolomic data focusing on recognizing combinations or "panels" of potential metabolic diagnostic biomarkers was evaluated. Utilizing LC/TOF MS, sera from 37 ovarian cancer patients and 35 benign controls were studied. Optimum panels of spectral features observed in positive or/and negative ion mode electrospray (ESI) MS with the ability to distinguish between control and ovarian cancer samples were selected using state-of-the-art feature selection methods such as recursive feature elimination and L1-norm SVM. Three evaluation processes (leave-one-out-cross-validation, 12-fold-cross-validation, 52-20-split-validation) were used to examine the SVM models based on the selected panels in terms of their ability for differentiating control vs. disease serum samples. The statistical significance for these feature selection results were comprehensively investigated. Classification of the serum sample test set was over 90% accurate indicating promise that the above approach may lead to the development of an accurate and reliable metabolomic-based approach for detecting ovarian cancer.
登录
查看更多内容
影响因子：
3
作者：
Furlanello C;Serafini M;Merler S;Jurman G
通讯作者：
Jurman G
影响因子：
--
作者：
Cherkassky, V
通讯作者：
Cherkassky, V
影响因子：
4.7
作者：
PETRU, E;SEVIN, BU;HILSENBECK, S
通讯作者：
HILSENBECK, S
影响因子：
2.4
作者：
Barker, M;Rayens, W
通讯作者：
Rayens, W
影响因子：
7.5
作者：
Li, LH;Tang, H;Clark, RA
通讯作者：
Clark, RA