Singular value decomposition for genome-wide expression data processing and modeling

We describe the use of singular value decomposition in transforming genome-wide expression data from genes × arrays space to reduced diagonalized “eigengenes” × “eigenarrays” space, where the eigengenes (or eigenarrays) are unique orthonormal superpositions of the genes (or arrays). Normalizing the data by filtering out the eigengenes (and eigenarrays) that are inferred to represent noise or experimental artifacts enables meaningful comparison of the expression of different genes across different arrays in different experiments. Sorting the data according to the eigengenes and eigenarrays gives a global picture of the dynamics of gene expression, in which individual genes and arrays appear to be classified into groups of similar regulation and function, or similar cellular state and biological phenotype, respectively. After normalization and sorting, the significant eigengenes and eigenarrays can be associated with observed genome-wide effects of regulators, or with measured samples, in which these regulators are overactive or underactive, respectively.

Dynamic regulation of the tryptophan operon: A modeling study and comparison with experimental data

A mathematical model for regulation of the tryptophan operon is presented. This model takes into account repression, feedback enzyme inhibition, and transcriptional attenuation. Special attention is given to model parameter estimation based on experimental data. The model's system of delay differential equations is numerically solved, and the results are compared with experimental data on the temporal evolution of enzyme activity in cultures of Escherichia coli after a nutritional shift (minimal + tryptophan medium to minimal medium). Good agreement is obtained between the numeric simulations and the experimental results for wild-type E. coli, as well as for two different mutant strains.

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.

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.

Blood flow and oxygen delivery to human brain during functional activity: Theoretical modeling and experimental data

Coupling of cerebral blood flow (CBF) and cerebral metabolic rate for oxygen (CMRO2) in physiologically activated brain states remains the subject of debates. Recently it was suggested that CBF is tightly coupled to oxidative metabolism in a nonlinear fashion. As part of this hypothesis, mathematical models of oxygen delivery to the brain have been described in which disproportionately large increases in CBF are necessary to sustain even small increases in CMRO2 during activation. We have explored the coupling of CBF and oxygen delivery by using two complementary methods. First, a more complex mathematical model was tested that differs from those recently described in that no assumptions were made regarding tissue oxygen level. Second, [15O] water CBF positron emission tomography (PET) studies in nine healthy subjects were conducted during states of visual activation and hypoxia to examine the relationship of CBF and oxygen delivery. In contrast to previous reports, our model showed adequate tissue levels of oxygen could be maintained without the need for increased CBF or oxygen delivery. Similarly, the PET studies demonstrated that the regional increase in CBF during visual activation was not affected by hypoxia. These findings strongly indicate that the increase in CBF associated with physiological activation is regulated by factors other than local requirements in oxygen.

The structure of the acto-myosin subfragment 1 complex: Results of searches using data from electron microscopy and x-ray crystallography

Surmises of how myosin subfragment 1 (S1) interacts with actin filaments in muscle contraction rest upon knowing the relative arrangement of the two proteins. Although there exist crystallographic structures for both S1 and actin, as well as electron microscopy data for the acto–S1 complex (AS1), modeling of this arrangement has so far only been done “by eye.” Here we report fitted AS1 structures obtained using a quantitative method that is both more objective and makes more complete use of the data. Using undistorted crystallographic results, the best-fit AS1 structure shows significant differences from that obtained by visual fitting. The best fit is produced using the F-actin model of Holmes et al. [Holmes, K. C., Popp, D., Gebhard, W. & Kabsch, W. (1990) Nature (London) 347, 44–49]. S1 residues at the AS1 interface are now found at a higher radius as well as being translated axially and rotated azimuthally. Fits using S1 plus loops missing from the crystal structure were achieved using a homology search method to predict loop structures. These improved fits favor an arrangement in which the loop at the 50- to 20-kDa domain junction of S1 is located near the N terminus of actin. Rigid-body movements of the lower 50-kDa domain, which further improve the fit, produce closure of the large 50-kDa domain cleft and bring conserved residues in the lower 50-kDa domain into an apparently appropriate orientation for close interaction with actin. This finding supports the idea that binding of ATP to AS1 at the end of the ATPase cycle disrupts the actin binding site by changing the conformation of the 50-kDa cleft of S1.

Data assimilation and its applications

In data assimilation, one prepares the grid data as the best possible estimate of the true initial state of a considered system by merging various measurements irregularly distributed in space and time, with a prior knowledge of the state given by a numerical model. Because it may improve forecasting or modeling and increase physical understanding of considered systems, data assimilation now plays a very important role in studies of atmospheric and oceanic problems. Here, three examples are presented to illustrate the use of new types of observations and the ability of improving forecasting or modeling.