Quantifying gene network connectivity in silico: scalability and accuracy of a modular approach.

Large, complex data sets that are generated from microarray experiments, create a need for systematic analysis techniques to unravel the underlying connectivity of gene regulatory networks. A modular approach, previously proposed by Kholodenko and co-workers, helps to scale down the network complexity into more computationally manageable entities called modules. A functional module includes a gene's mRNA, promoter and resulting products, thus encompassing a large set of interacting states. The essential elements of this approach are described in detail for a three-gene model network and later extended to a ten-gene model network, demonstrating scalability. The network architecture is identified by analysing in silico steady-state changes in the activities of only the module outputs, communicating intermediates, that result from specific perturbations applied to the network modules one at a time. These steady-state changes form the system response matrix, which is used to compute the network connectivity or network interaction map. By employing a known biochemical network, the accuracy of the modular approach and its sensitivity to key assumptions are evaluated.

Automated quantification of quantum-dot-labelled epidermal growth factor receptor internalization via multiscale image segmentation.

The ability to monitor epidermal growth factor receptor (EGFr) internalization specifically, and cellular protein concentrations and activation states in general, has been recently improved by the use of appropriately functionalized quantum dots (QDs), as a result of the long-lasting fluorescence, brightness and multicolour of these nanoparticles. However, important quantitative information about locational proteomics is based on the analysis of the properties of many cells and cell cultures on a per-cell basis, rather than tracking individual events within one cell. Moreover, relative positional information is often gained from traditional staining protocols of distinct cellular compartments that are prone to noise, fading and low contrast. We apply a novel multiscale image segmentation based on region growing to classify automatically objects in fixed cell preparations and to define regional zones in all cells prior to QD concentration measures. This allows rapid quantitative description of EGFr internalization as it changes with incubation time. The capabilities realizable by simultaneous application of confocal imaging and functionalized QDs in conjunction with advanced image analysis are a prerequisite for automated and multiplexed cytomics assays.

Automated tissue analysis--a bioinformatics perspective.

OBJECTIVES: Recent progress in automated tissue analysis (tissomics) provides reproducible phenotypical characterization of histological specimens. We introduce informatics tools to cluster and correlate quantitative tissue profiles with gene expression data. The great potential of synergies between tissue analysis and bioinformatics and its perspectives in medical research and computational diagnostics are discussed. METHODS: Key enablers in microscopic imaging and machine vision are reviewed to perform a high-throughput tissue analysis. Methodologies are described and results are demonstrated that support a combined analysis of tissue with gene expression profiles whereby the consideration of individual responses is key. RESULTS: Comprehensive histomorphometric profiles, extracted using machine vision, provide information regarding the components and heterogeneity of a tissue in a reproducible format amenable to data mining and analysis. Tissue quantitative information can be placed in synergetic context with bioinformatics data, such as gene expression profiles, for a more comprehensive stratification of individual responses. From a bioinformatics point of view tissue data are co-variants that support the identification of candidate genes relevant in tissue injury or disease. CONCLUSIONS: Progress in automated analytics enables the generation of quantitative data about tissue previously limited to visual histopathology. Such reproducible data sets can be statistically correlated and clustered throughout the continuum of bioinformatics. The combined approach supports a system-wide view of biology and has a potential to accelerate developments for a personalized computational diagnosis.

Analysis of gene expression in carbon tetrachloride-treated rat livers using a novel bioarray technology.

The present study successfully utilizes a new ADME Rat Expression Bioarray, containing 1040 metabolism- and toxicology-linked genes, to monitor gene expression from the livers of rats treated with carbon tetrachloride (CCl(4)). Histopathological analysis, hierarchical clustering methods, and gene expression profiling are compared between the control and CCl(4)-treated animals. A total of 44 transcripts were found to be altered in response to the hepatotoxin, 19 of which were upregulated and 25 were downregulated. Some of these gene expression changes were expected and concurred with previously published data while others were novel findings.

Quality measures in applications of image restoration.

We describe a new method for the estimation of image quality in image restoration applications. We demonstrate this technique on a simulated data set of fluorescent beads, in comparison with restoration by three different deconvolution methods. Both the number of iterations and a regularisation factor are varied to enforce changes in the resulting image quality. First, the data sets are directly compared by an accuracy measure. These values serve to validate the image quality descriptor, which is developed on the basis of optical information theory. This most general measure takes into account the spectral energies and the noise, weighted in a logarithmic fashion. It is demonstrated that this method is particularly helpful as a user-oriented method to control the output of iterative image restorations and to eliminate the guesswork in choosing a suitable number of iterations.

Form and function of mammalian lung: analysis by scientific computing.

This study approaches the investigation of airway morphology of the lung with a new set of imaging and computer graphical methods, including confocal imaging, computer-guided image acquisition, visualization and fractal graphics. The key result is that, in contrast to the belief that the design of the conductive part of lung of smaller mammals can be described with a trumpet model, the findings reported here document a strongly monopodial branching pattern with the functional consequence of a variation of dead space between the trachea and the acini. This non-dichotomic structural design finds its continuation within the respiratory units as the necessary requirement for an optimal space filling and dense packing which cannot be achieved by a dichotomic branching only. Based on a computer model, computational physics tightly coupled with computer visualistics enables functional simulation of the lung model regarding gas transport. The predicted variance in the ventilation of acini gives rise to an explanation of the well-known difference between the morphologically predicted and physiologically required diffusion capacity.

Intraoperative three-dimensional ultrasonography: an approach to register brain shift using multidimensional image processing.

Neuronavigation uses the skull as a reference system for transfer of image-space data to physical space during brain surgery. This requires a stable spatial relation between the skull and intracranial structures. However, especially dura opening and preparation for lesion removal causes brain shift. This shift may mislead the surgeon unless preoperatively defined image-space data are corrected for shifting online intraoperatively. Since a real-time modality is required intraoperatively, we propose three-dimensional (3 D) ultrasonography for detection of brain shift. By coupling common ultrasound probes (3.5/6.5 MHz) to a magnetic digitizer receiver 2 D-ultrasound scans were obtained intraoperatively along with their spatial orientation. 3 D-ultrasonography was achieved by alignment of sequentially obtained 2 D-scans. For multimodal matching, preoperative MRI data was segmented for landmarks (cerebral ventricles, lesion) automatically. The 3 D-ultrasonography data set scanned intraoperatively was contoured and matched with the MRI data set. Intraoperative 3 D-ultrasonography revealed excellent delineation of landmarks in almost real time in six patients studied. Matching of MRI data and intraoperative 3 D-ultrasonography data was successful with good correspondence of landmarks. Intraoperative 3 D-ultrasonography is proposed as a promising tool for on-line detection of brain shift during intracranial operations.

3-D confocal laser scanning microscopy used in morphometric analysis of rat Purkinje cell dendritic spines after chronic ethanol consumption.

A confocal laser scanning microscope (with a 543 nm laser) was used for imaging rat Purkinje cell dendritic spines at high 3-D resolution. In a nutritionally controlled study of the rat, 5 months of ethanol consumption was demonstrated to alter the spines of Purkinje cell dendrites in rat cerebellum. Intact spines showed significant elongation after ethanol exposure, whereas this neuromorphological alteration could not be detected in controls. Spine elongation could be regarded as compensative growth of spines in search of new synaptic contacts due to alcohol induced cell loss.

3-D TOP-A software package for the topological analysis of image sequences.

The software package 3-D TOP is a measurement software which has been developed to meet the needs of those who analyze mechanical serial sections in histology and optical sections in laser scanning microscopy. The semiautomatic definition of topological centerlines is the backbone of this software. The three basic advantages in using the concept of topology are (a) resolving ambiguities for surface rendering, (b) flexible definition of objects for automatic identification, labeling, and measurement, and (c) topological quantification of complex forms. A specific application of this software to lung research is demonstrated.

Dental anatomy portrayed with microscopic volume investigations.

The clinical treatment of the root canal of teeth--called endodontics--assumes a precise idea of the spatial arrangement of the anatomy of teeth and their inner structure. By using computer-assisted data acquisition from filmed sequences of histologic serial sections and a special kind of magnetic resonance microscope--the Stray Field Imaging (STRAFI)--volume investigations were carried out using special functions of a newly developed 3D software. Possible applications and future perspectives are discussed.

A method for spatio-temporal (4-D) data representation in confocal microscopy: application to neuroanatomical plasticity.

This paper describes a new method for data representation and visualization in four dimensions (three dimensions plus time). Sequential volumes, exhibiting morphological activity, are acquired non-invasively with a confocal scanning laser microscope, where each data set corresponds to a time sample. A pipelined processing includes packing of volumes and specific volume rendering techniques. Subsequent processing in HIS (hue, intensity, saturation)l colour space combines functional, packed images with shaded three-dimensional views. As a result, even subtle changes in morphology become visible and computational time is saved. Experimental findings obtained from investigations of synaptic plasticity in cultured retinal tissue are reported.

Ultraviolet confocal fluorescence microscopy of the in vitro cornea: redox metabolic imaging.

A laser scanning microscope was fitted with two argon-ion lasers that provided wavelengths in the regions of 364, 488, and 514 nm. A Zeiss water objective of 25 x , with a numerical aperture of 0.8, corrected for the UV, was used to measure the fluorescence from optical sections of freshly enucleated rabbit eyes. The confocal microscope was used in both the reflected and fluorescent modes to image in situ epithelial and endothelial cells. An excitation wavelength of 364 nm and emission at 400-500 nm were used to image the fluorescence from reduced pyridine nucleotides. We demonstrate the feasibility of two-dimensional fluorescent confocal imaging of reduced pyridine nucleotides in corneal epithelial and endothelial cells.

Methodical aspects of 3-D reconstruction of chromatin architecture in mouse trophoblast giant nuclei.

Precise 3-D reconstruction of the spatial organization of murine trophoblast giant-cell chromatin is a prerequisite for detailed investigations on the fine structural changes in chromatin-fibre organization during the trophoblast endomitotic cell cycle. It appears very likely that sequential fine structural changes in the chromatin arrangement are concomitant with the changes in the gene expression pattern of these cells during the early phase of murine gestation. The complex intra-nuclear chromatin organization of mouse trophoblast giant nuclei was investigated in permanent tissue preparations which had been stained with a DNA-specific dye. The spatial chromatin arrangement was examined in fluorescence with a confocal scanning laser microscope. Serial optical sections were recorded and subjected to a computer-based 3-D reconstruction method which is suitable even for very complex biological structures.

Computerized three-dimensional reconstructions of serial sections in electron microscopy.

To improve reconstructive 3D electron microscopy novel methods are discussed to represent and process serial section images in a cuberille environment. This includes the analysis of the transfer characteristics of the image detection system, the use of laser-induced fiducials for deformation correction and alignment, the control of section thickness by EELS and the use of ESI to image thick sections.

[Differential diagnosis of papillary carcinomas of the thyroid, using image analysis and three dimensional reconstruction from serial sections]. / Differentialdiagnosen des papillären Karzinoms der Schilddrüse mittels morphometrischer Bildanalyse und dreidimensionaler Geweberekonstruktion.

Papillae with fibrovascular cores are characteristic of papillary carcinoma of the thyroid. Papillae may be found in diffuse hyperplasia, nodular hyperplasia, Hashimoto's disease and follicular adenoma. Tissues from ten benign hyperplasias and ten papillary carcinomas were reconstructed from serial sections with three dimensional reconstruction programs. Significant qualitative and quantitative differences were found between the hyperplasia and the carcinoma. The principal differences between papillae of papillary carcinoma and hyperplasia were more clearly seen in the three dimensional reconstruction, than by means of morphometric methods. Certain criteria, e.g. the volume of papillae, were useful only with regard to the third dimension. Nevertheless, three dimensional reconstruction of biological tissue is a time consuming procedure which is not yet suitable for routine examination.

The serotonin-immunoreactive system of the suprachiasmatic nucleus in the hibernating ground squirrel, Spermophilus richardsonii.

In the suprachiasmatic nucleus (NSC) of hibernating and non-hibernating ground squirrels, the distribution of serotonin-immunoreactive (5HT-IR) fibers was studied by the use of the peroxidase-antiperoxidase technique. The cytology of perikarya giving rise to these suprachiasmatic 5HT-IR fibers was investigated in the anterior raphe nuclei. Differences in the immunoreactivity of suprachiasmatic fibers between hibernating and non-hibernating ground squirrels were determined by digital image analysis. The cellular activity was determined by digital image analysis. The cellular activity was determined densitometrically after RNA-staining in anterior raphe neurons and suprachiasmatic perikarya. Abundant 5HT-IR fibers were observed in the medial and ventromedial portions of the NSC. Frequently, the fibers were found in close contact with perikarya of suprachiasmatic neurons. The central portion of the nucleus and the surrounding hypothalamic areas contained only a few scattered 5HT-IR fibers. Inside the raphe nuclei, 5HT-IR fibers and perikarya formed a dense network. In hibernating ground squirrels, the immunoreactivity to serotonin was approximately 45% higher than in non-hibernating controls. This difference is in accordance with signs of higher neuronal activity (40% higher RNA-content, 20% larger cell nuclei) in 5HT-IR perikarya of the raphe nucleus and the persisting activity of the NSC during hibernation; the activity of other brain regions dropped conspicuously in torpid animals.

On-line transmission electron microscopic image analysis of chromatin texture for differentiation of thyroid gland tumors.

Nuclei of the cells from the thyroid gland were analyzed in a transmission electron microscope by direct TV scanning and on-line image processing. The method uses the advantages of a visual-perception model to detect structures in noisy and low-contrast images. The features analyzed include area, a form factor and texture parameters from the second derivative stage. Three tumor-free thyroid tissues, three follicular adenomas, three follicular carcinomas and three papillary carcinomas were studied. The computer-aided cytophotometric method showed that the most significant differences were the statistics of the chromatin texture features of homogeneity and regularity. These findings document the possibility of an automated differentiation of tumors at the ultrastructural level.

Computer-based cytophotometric classification of thyroid tumors in imprints.

This study of 14 follicular adenomas, 10 papillary carcinomas and 11 follicular carcinomas of the human thyroid gland demonstrates the possibility of a cytological tumor classification using digital picture processing. Routinely prepared, HE-stained imprints of surgical specimens were scanned under a light microscope at high resolution with a colour TV camera. The cell nuclei were segmented and analysed with an image processing system. The computer-aided cytophotometric methods detected the most significant differences in the chromatin texture with a criteria variance of texture line distances and texture points per texture knots. Using these criteria benign and malignant tumor types could be successfully differentiated.

Computer analysis of chromatin arrangement and nuclear texture in follicular thyroid tumours.

The differentiation of the thyroid glands follicular neoplasias into adenomas and carcinomas is currently done using the histological criteria recommended by WHO. This pilot study of 10 human follicular carcinomas and 10 folliculars adenomas demonstrates the possibility of a cytological classification using digital picture processing of high resolution cell images. Giemsa stained paraplast sections were scanned with a Colour-TV-camera, different channels were used with respect to staining and analyzing methods and computed with an image processing system. The computer aided cytophotometric methods detected significant differences in the chromatin arrangement and structure.