Recursive partitioning for tumor classification with gene expression microarray data

Precise classification of tumors is critically important for cancer diagnosis and treatment. It is also a scientifically challenging task. Recently, efforts have been made to use gene expression profiles to improve the precision of classification, with limited success. Using a published data set for purposes of comparison, we introduce a methodology based on classification trees and demonstrate that it is significantly more accurate for discriminating among distinct colon cancer tissues than other statistical approaches used heretofore. In addition, competing classification trees are displayed, which suggest that different genes may coregulate colon cancers.

Statistical modeling of large microarray data sets to identify stimulus-response profiles

A statistical modeling approach is proposed for use in searching large microarray data sets for genes that have a transcriptional response to a stimulus. The approach is unrestricted with respect to the timing, magnitude or duration of the response, or the overall abundance of the transcript. The statistical model makes an accommodation for systematic heterogeneity in expression levels. Corresponding data analyses provide gene-specific information, and the approach provides a means for evaluating the statistical significance of such information. To illustrate this strategy we have derived a model to depict the profile expected for a periodically transcribed gene and used it to look for budding yeast transcripts that adhere to this profile. Using objective criteria, this method identifies 81% of the known periodic transcripts and 1,088 genes, which show significant periodicity in at least one of the three data sets analyzed. However, only one-quarter of these genes show significant oscillations in at least two data sets and can be classified as periodic with high confidence. The method provides estimates of the mean activation and deactivation times, induced and basal expression levels, and statistical measures of the precision of these estimates for each periodic transcript.

The Stanford Microarray Database

The Stanford Microarray Database (SMD) stores raw and normalized data from microarray experiments, and provides web interfaces for researchers to retrieve, analyze and visualize their data. The two immediate goals for SMD are to serve as a storage site for microarray data from ongoing research at Stanford University, and to facilitate the public dissemination of that data once published, or released by the researcher. Of paramount importance is the connection of microarray data with the biological data that pertains to the DNA deposited on the microarray (genes, clones etc.). SMD makes use of many public resources to connect expression information to the relevant biology, including SGD [Ball,C.A., Dolinski,K., Dwight,S.S., Harris,M.A., Issel-Tarver,L., Kasarskis,A., Scafe,C.R., Sherlock,G., Binkley,G., Jin,H. et al. (2000) Nucleic Acids Res., 28, 77–80], YPD and WormPD [Costanzo,M.C., Hogan,J.D., Cusick,M.E., Davis,B.P., Fancher,A.M., Hodges,P.E., Kondu,P., Lengieza,C., Lew-Smith,J.E., Lingner,C. et al. (2000) Nucleic Acids Res., 28, 73–76], Unigene [Wheeler,D.L., Chappey,C., Lash,A.E., Leipe,D.D., Madden,T.L., Schuler,G.D., Tatusova,T.A. and Rapp,B.A. (2000) Nucleic Acids Res., 28, 10–14], dbEST [Boguski,M.S., Lowe,T.M. and Tolstoshev,C.M. (1993) Nature Genet., 4, 332–333] and SWISS-PROT [Bairoch,A. and Apweiler,R. (2000) Nucleic Acids Res., 28, 45–48] and can be accessed at http://genome-www.stanford.edu/microarray.

Amygdala-enriched genes identified by microarray technology are restricted to specific amygdaloid subnuclei

Microarray technology represents a potentially powerful method for identifying cell type- and regionally restricted genes expressed in the brain. Here we have combined a microarray analysis of differential gene expression among five selected brain regions, including the amygdala, cerebellum, hippocampus, olfactory bulb, and periaqueductal gray, with in situ hybridization. On average, 0.3% of the 34,000 genes interrogated were highly enriched in each of the five regions, relative to the others. In situ hybridization performed on a subset of amygdala-enriched genes confirmed in most cases the overall region-specificity predicted by the microarray data and identified additional sites of brain expression not examined on the microarrays. Strikingly, the majority of these genes exhibited boundaries of expression within the amygdala corresponding to cytoarchitectonically defined subnuclei. These results define a unique set of molecular markers for amygdaloid subnuclei and provide tools to genetically dissect their functional roles in different emotional behaviors.

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.

Neural-network-based classification of cognitively normal, demented, Alzheimer disease and vascular dementia from single photon emission with computed tomography image data from brain.

Single photon emission with computed tomography (SPECT) hexamethylphenylethyleneamineoxime technetium-99 images were analyzed by an optimal interpolative neural network (OINN) algorithm to determine whether the network could discriminate among clinically diagnosed groups of elderly normal, Alzheimer disease (AD), and vascular dementia (VD) subjects. After initial image preprocessing and registration, image features were obtained that were representative of the mean regional tissue uptake. These features were extracted from a given image by averaging the intensities over various regions defined by suitable masks. After training, the network classified independent trials of patients whose clinical diagnoses conformed to published criteria for probable AD or probable/possible VD. For the SPECT data used in the current tests, the OINN agreement was 80 and 86% for probable AD and probable/possible VD, respectively. These results suggest that artificial neural network methods offer potential in diagnoses from brain images and possibly in other areas of scientific research where complex patterns of data may have scientifically meaningful groupings that are not easily identifiable by the researcher.

Comprehensive Identification of Cell Cycle–regulated Genes of the Yeast Saccharomyces cerevisiae by Microarray HybridizationD⃞

We sought to create a comprehensive catalog of yeast genes whose transcript levels vary periodically within the cell cycle. To this end, we used DNA microarrays and samples from yeast cultures synchronized by three independent methods: α factor arrest, elutriation, and arrest of a cdc15 temperature-sensitive mutant. Using periodicity and correlation algorithms, we identified 800 genes that meet an objective minimum criterion for cell cycle regulation. In separate experiments, designed to examine the effects of inducing either the G1 cyclin Cln3p or the B-type cyclin Clb2p, we found that the mRNA levels of more than half of these 800 genes respond to one or both of these cyclins. Furthermore, we analyzed our set of cell cycle–regulated genes for known and new promoter elements and show that several known elements (or variations thereof) contain information predictive of cell cycle regulation. A full description and complete data sets are available at http://cellcycle-www.stanford.edu

Importance of replication in microarray gene expression studies: Statistical methods and evidence from repetitive cDNA hybridizations

We present statistical methods for analyzing replicated cDNA microarray expression data and report the results of a controlled experiment. The study was conducted to investigate inherent variability in gene expression data and the extent to which replication in an experiment produces more consistent and reliable findings. We introduce a statistical model to describe the probability that mRNA is contained in the target sample tissue, converted to probe, and ultimately detected on the slide. We also introduce a method to analyze the combined data from all replicates. Of the 288 genes considered in this controlled experiment, 32 would be expected to produce strong hybridization signals because of the known presence of repetitive sequences within them. Results based on individual replicates, however, show that there are 55, 36, and 58 highly expressed genes in replicates 1, 2, and 3, respectively. On the other hand, an analysis by using the combined data from all 3 replicates reveals that only 2 of the 288 genes are incorrectly classified as expressed. Our experiment shows that any single microarray output is subject to substantial variability. By pooling data from replicates, we can provide a more reliable analysis of gene expression data. Therefore, we conclude that designing experiments with replications will greatly reduce misclassification rates. We recommend that at least three replicates be used in designing experiments by using cDNA microarrays, particularly when gene expression data from single specimens are being analyzed.

Will new diagnostic criteria for diabetes mellitus change phenotype of patients with diabetes? Reanalysis of European epidemiological data

Objective: To evaluate the impact of the revised diagnostic criteria for diabetes mellitus adopted by the American Diabetes Association on prevalence of diabetes and on classification of patients. For epidemiological purposes the American criteria use a fasting plasma glucose concentration ⩾7.0 mmol/l in contrast with the current World Health Organisation criteria of 2 hour glucose concentration ⩾11.1 mmol/l.

Dynamic modeling of gene expression data

We describe the time evolution of gene expression levels by using a time translational matrix to predict future expression levels of genes based on their expression levels at some initial time. We deduce the time translational matrix for previously published DNA microarray gene expression data sets by modeling them within a linear framework by using the characteristic modes obtained by singular value decomposition. The resulting time translation matrix provides a measure of the relationships among the modes and governs their time evolution. We show that a truncated matrix linking just a few modes is a good approximation of the full time translation matrix. This finding suggests that the number of essential connections among the genes is small.

A rapid classification protocol for the CATH Domain Database to support structural genomics

In order to support the structural genomic initiatives, both by rapidly classifying newly determined structures and by suggesting suitable targets for structure determination, we have recently developed several new protocols for classifying structures in the CATH domain database (http://www.biochem.ucl.ac.uk/bsm/cath). These aim to increase the speed of classification of new structures using fast algorithms for structure comparison (GRATH) and to improve the sensitivity in recognising distant structural relatives by incorporating sequence information from relatives in the genomes (DomainFinder). In order to ensure the integrity of the database given the expected increase in data, the CATH Protein Family Database (CATH-PFDB), which currently includes 25 320 structural domains and a further 160 000 sequence relatives has now been installed in a relational ORACLE database. This was essential for developing more rigorous validation procedures and for allowing efficient querying of the database, particularly for genome analysis. The associated Dictionary of Homologous Superfamilies [Bray,J.E., Todd,A.E., Pearl,F.M.G., Thornton,J.M. and Orengo,C.A. (2000) Protein Eng., 13, 153–165], which provides multiple structural alignments and functional information to assist in assigning new relatives, has also been expanded recently and now includes information for 903 homo­logous superfamilies. In order to improve coverage of known structures, preliminary classification levels are now provided for new structures at interim stages in the classification protocol. Since a large proportion of new structures can be rapidly classified using profile-based sequence analysis [e.g. PSI-BLAST: Altschul,S.F., Madden,T.L., Schaffer,A.A., Zhang,J., Zhang,Z., Miller,W. and Lipman,D.J. (1997) Nucleic Acids Res., 25, 3389–3402], this provides preliminary classification for easily recognisable homologues, which in the latest release of CATH (version 1.7) represented nearly three-quarters of the non-identical structures.

PDB-REPRDB: a database of representative protein chains from the Protein Data Bank (PDB)

PDB-REPRDB is a database of representative protein chains from the Protein Data Bank (PDB). The previous version of PDB-REPRDB provided 48 representative sets, whose similarity criteria were predetermined, on the WWW. The current version is designed so that the user may obtain a quick selection of representative chains from PDB. The selection of representative chains can be dynamically configured according to the user’s requirement. The WWW interface provides a large degree of freedom in setting parameters, such as cut-off scores of sequence and structural similarity. One can obtain a representative list and classification data of protein chains from the system. The current database includes 20 457 protein chains from PDB entries (August 6, 2000). The system for PDB-REPRDB is available at the Parallel Protein Information Analysis system (PAPIA) WWW server (http://www.rwcp.or.jp/papia/).

Saccharomyces Genome Database provides tools to survey gene expression and functional analysis data

Upon the completion of the Saccharomyces cerevisiae genomic sequence in 1996 [Goffeau,A. et al. (1997) Nature, 387, 5], several creative and ambitious projects have been initiated to explore the functions of gene products or gene expression on a genome-wide scale. To help researchers take advantage of these projects, the Saccharomyces Genome Database (SGD) has created two new tools, Function Junction and Expression Connection. Together, the tools form a central resource for querying multiple large-scale analysis projects for data about individual genes. Function Junction provides information from diverse projects that shed light on the role a gene product plays in the cell, while Expression Connection delivers information produced by the ever-increasing number of microarray projects. WWW access to SGD is available at genome-www.stanford.edu/Saccharomyces/.

Protein Information Resource: a community resource for expert annotation of protein data

The Protein Information Resource, in collaboration with the Munich Information Center for Protein Sequences (MIPS) and the Japan International Protein Information Database (JIPID), produces the most comprehensive and expertly annotated protein sequence database in the public domain, the PIR-International Protein Sequence Database. To provide timely and high quality annotation and promote database interoperability, the PIR-International employs rule-based and classification-driven procedures based on controlled vocabulary and standard nomenclature and includes status tags to distinguish experimentally determined from predicted protein features. The database contains about 200 000 non-redundant protein sequences, which are classified into families and superfamilies and their domains and motifs identified. Entries are extensively cross-referenced to other sequence, classification, genome, structure and activity databases. The PIR web site features search engines that use sequence similarity and database annotation to facilitate the analysis and functional identification of proteins. The PIR-Inter­national databases and search tools are accessible on the PIR web site at http://pir.georgetown.edu/ and at the MIPS web site at http://www.mips.biochem.mpg.de. The PIR-International Protein Sequence Database and other files are also available by FTP.

A fully automatic evolutionary classification of protein folds: Dali Domain Dictionary version 3

The Dali Domain Dictionary (http://www.ebi.ac.uk/dali/domain) is a numerical taxonomy of all known structures in the Protein Data Bank (PDB). The taxonomy is derived fully automatically from measurements of structural, functional and sequence similarities. Here, we report the extension of the classification to match the traditional four hierarchical levels corresponding to: (i) supersecondary structural motifs (attractors in fold space), (ii) the topology of globular domains (fold types), (iii) remote homologues (functional families) and (iv) homologues with sequence identity above 25% (sequence families). The computational definitions of attractors and functional families are new. In September 2000, the Dali classification contained 10 531 PDB entries comprising 17 101 chains, which were partitioned into five attractor regions, 1375 fold types, 2582 functional families and 3724 domain sequence families. Sequence families were further associated with 99 582 unique homologous sequences in the HSSP database, which increases the number of effectively known structures several-fold. The resulting database contains the description of protein domain architecture, the definition of structural neighbours around each known structure, the definition of structurally conserved cores and a comprehensive library of explicit multiple alignments of distantly related protein families.