Clustering with position-specific constraints on variance: applying redescending M-estimators to label-free LC-MS data analysis.
Clustering with position-specific constraints on variance: applying redescending M-estimators to label-free LC-MS data analysis.
复制标题
DOI：
10.1186/1471-2105-12-358
复制
发表时间：
2011-08-31
影响因子：
3
通讯作者：
Pyne S
中科院分区：
文献类型：
--
作者：
Frühwirth R;Mani DR;Pyne S
聚类是一种广泛应用的模式识别方法，用于发现数据中的相似观测组。虽然有许多不同的聚类算法，但很少有算法能够对包含在给定聚类中的数据点的属性变化施加约束。特别是，可以根据簇的位置(质心位置)限制簇内变化的簇算法可以在许多重要应用中产生有效和优化的结果，从高能物理中的硅像素或量热单元的簇到蛋白质组学和代谢组学中的基于无标记的液质联用(LC-MS)数据分析。提出了一种基于M-估计器的无监督算法MEDEA(M-Estimator with Desitive Anneation)，该算法旨在对聚类过程中的方差施加特定位置的约束。通过将MEDEA应用于蛋白质组生物标记物发现中的“峰匹配”问题--识别多个样品中的公共LC-MS峰，证明了MEDEA的实用性。使用真实的数据集，我们表明MEDEA不仅比目前最先进的基于模型的聚类方法性能更好，而且实现的效率要高得多，因此适用于更大的LC-MS数据集。MEDEA是解决无标记LC-MS数据中峰匹配问题的有效方法。实施MEDEA算法的程序，包括数据集、聚类结果和补充信息，可从作者网站http://www.hephy.at/user/fru/medea/.获得
Clustering is a widely applicable pattern recognition method for discovering groups of similar observations in data. While there are a large variety of clustering algorithms, very few of these can enforce constraints on the variation of attributes for data points included in a given cluster. In particular, a clustering algorithm that can limit variation within a cluster according to that cluster's position (centroid location) can produce effective and optimal results in many important applications ranging from clustering of silicon pixels or calorimeter cells in high-energy physics to label-free liquid chromatography based mass spectrometry (LC-MS) data analysis in proteomics and metabolomics. We present MEDEA (M-Estimator with DEterministic Annealing), an M-estimator based, new unsupervised algorithm that is designed to enforce position-specific constraints on variance during the clustering process. The utility of MEDEA is demonstrated by applying it to the problem of "peak matching"--identifying the common LC-MS peaks across multiple samples--in proteomic biomarker discovery. Using real-life datasets, we show that MEDEA not only outperforms current state-of-the-art model-based clustering methods, but also results in an implementation that is significantly more efficient, and hence applicable to much larger LC-MS data sets. MEDEA is an effective and efficient solution to the problem of peak matching in label-free LC-MS data. The program implementing the MEDEA algorithm, including datasets, clustering results, and supplementary information is available from the author website at http://www.hephy.at/user/fru/medea/.
登录
查看更多内容
影响因子：
5.8
作者：
Bellew, Matthew;Coram, Marc;McIntosh, Martin
通讯作者：
McIntosh, Martin
影响因子：
7
作者：
Olsen, JV;de Godoy, LMF;Mann, M
通讯作者：
Mann, M
影响因子：
1.9
作者：
BANFIELD, JD;RAFTERY, AE
通讯作者：
RAFTERY, AE
影响因子：
8
作者：
Li, SZ
通讯作者：
Li, SZ
影响因子：
7
作者：
Jaffe, Jacob D.;Mani, D. R.;Carr, Steven A.
通讯作者：
Carr, Steven A.