Knowledge-based analysis of microarray gene expression data by using support vector machines

We introduce a method of functionally classifying genes by using gene expression data from DNA microarray hybridization experiments. The method is based on the theory of support vector machines (SVMs). SVMs are considered a supervised computer learning method because they exploit prior knowledge of gene function to identify unknown genes of similar function from expression data. SVMs avoid several problems associated with unsupervised clustering methods, such as hierarchical clustering and self-organizing maps. SVMs have many mathematical features that make them attractive for gene expression analysis, including their flexibility in choosing a similarity function, sparseness of solution when dealing with large data sets, the ability to handle large feature spaces, and the ability to identify outliers. We test several SVMs that use different similarity metrics, as well as some other supervised learning methods, and find that the SVMs best identify sets of genes with a common function using expression data. Finally, we use SVMs to predict functional roles for uncharacterized yeast ORFs based on their expression data.

Preserved neuron number in the hippocampus of aged rats with spatial learning deficits.

Hippocampal neuron loss is widely viewed as a hallmark of normal aging. Moreover, neuronal degeneration is thought to contribute directly to age-related deficits in learning and memory supported by the hippocampus. By taking advantage of improved methods for quantifying neuron number, the present study reports evidence challenging these long-standing concepts. The status of hippocampal-dependent spatial learning was evaluated in young and aged Long-Evans rats using the Morris water maze, and the total number of neurons in the principal cell layers of the dentate gyrus and hippocampus was quantified according to the optical fractionator technique. For each of the hippocampal fields, neuron number was preserved in the aged subjects as a group and in aged individuals with documented learning and memory deficits indicative of hippocampal dysfunction. The findings demonstrate that hippocampal neuronal degeneration is not an inevitable consequence of normal aging and that a loss of principal neurons in the hippocampus fails to account for age-related learning and memory impairment. The observed preservation of neuron number represents an essential foundation for identifying the neurobiological effects of hippocampal aging that account for cognitive decline.

Structure-activity relationships derived by machine learning: the use of atoms and their bond connectivities to predict mutagenicity by inductive logic programming.

We present a general approach to forming structure-activity relationships (SARs). This approach is based on representing chemical structure by atoms and their bond connectivities in combination with the inductive logic programming (ILP) algorithm PROGOL. Existing SAR methods describe chemical structure by using attributes which are general properties of an object. It is not possible to map chemical structure directly to attribute-based descriptions, as such descriptions have no internal organization. A more natural and general way to describe chemical structure is to use a relational description, where the internal construction of the description maps that of the object described. Our atom and bond connectivities representation is a relational description. ILP algorithms can form SARs with relational descriptions. We have tested the relational approach by investigating the SARs of 230 aromatic and heteroaromatic nitro compounds. These compounds had been split previously into two subsets, 188 compounds that were amenable to regression and 42 that were not. For the 188 compounds, a SAR was found that was as accurate as the best statistical or neural network-generated SARs. The PROGOL SAR has the advantages that it did not need the use of any indicator variables handcrafted by an expert, and the generated rules were easily comprehensible. For the 42 compounds, PROGOL formed a SAR that was significantly (P < 0.025) more accurate than linear regression, quadratic regression, and back-propagation. This SAR is based on an automatically generated structural alert for mutagenicity.