Search PubMed⌕ Search

Biomedical subjects

Darrin P Lewis

Publications and source records attributed to Darrin P Lewis.

3 recordsLinked to original sources

Support vector machine learning from heterogeneous data: an empirical analysis using protein sequence and structure.

MOTIVATION: Drawing inferences from large, heterogeneous sets of biological data requires a theoretical framework that is capable of representing, e.g. DNA and protein sequences, protein structures, microarray expression data, various types of interaction networks, etc. Recently, a class of algorithms known as kernel methods has emerged as a powerful framework for combining diverse types of data. The support vector machine (SVM) algorithm is the most popular kernel method, due to its theoretical underpinnings and strong empirical performance on a wide variety of classification tasks. Furthermore, several recently described extensions allow the SVM to assign relative weights to various datasets, depending upon their utilities in performing a given classification task. RESULTS: In this work, we empirically investigate the performance of the SVM on the task of inferring gene functional annotations from a combination of protein sequence and structure data. Our results suggest that the SVM is quite robust to noise in the input datasets. Consequently, in the presence of only two types of data, an SVM trained from an unweighted combination of datasets performs as well or better than a more sophisticated algorithm that assigns weights to individual data types. Indeed, for this simple case, we can demonstrate empirically that no solution is significantly better than the naive, unweighted average of the two datasets. On the other hand, when multiple noisy datasets are included in the experiment, then the naive approach fares worse than the weighted approach. Our results suggest that for many applications, a naive unweighted sum of kernels may be sufficient. AVAILABILITY: http://noble.gs.washington.edu/proj/seqstruct

Algorithms↗

Kernel hierarchical gene clustering from microarray expression data.

MOTIVATION: Unsupervised analysis of microarray gene expression data attempts to find biologically significant patterns within a given collection of expression measurements. For example, hierarchical clustering can be applied to expression profiles of genes across multiple experiments, identifying groups of genes that share similar expression profiles. Previous work using the support vector machine supervised learning algorithm with microarray data suggests that higher-order features, such as pairwise and tertiary correlations across multiple experiments, may provide significant benefit in learning to recognize classes of co-expressed genes. RESULTS: We describe a generalization of the hierarchical clustering algorithm that efficiently incorporates these higher-order features by using a kernel function to map the data into a high-dimensional feature space. We then evaluate the utility of the kernel hierarchical clustering algorithm using both internal and external validation. The experiments demonstrate that the kernel representation itself is insufficient to provide improved clustering performance. We conclude that mapping gene expression data into a high-dimensional feature space is only a good idea when combined with a learning algorithm, such as the support vector machine that does not suffer from the curse of dimensionality. AVAILABILITY: Supplementary data at www.cs.columbia.edu/compbio/hiclust. Software source code available by request.

Algorithms↗

Exploring gene expression data with class scores.

We address a commonly asked question about gene expression data sets: "What functional classes of genes are most interesting in the data?" In the methods we present, expression data is partitioned into classes based on existing annotation schemes. Each class is then given three separately derived "interest" scores. The first score is based on an assessment of the statistical significance of gene expression changes experienced by members of the class, in the context of the experimental design. The second is based on the co-expression of genes in the class. The third score is based on the learnability of the classification. We show that all three methods reveal significant classes in each of three different gene expression data sets. Many classes are identified by one method but not the others, indicating that the methods are complementary. The classes identified are in many cases of clear relevance to the experiment. Our results suggest that these class scoring methods are useful tools for exploring gene expression data.

Analysis of Variance↗