Quantitative comparison of taxa and taxon concepts in the diatom genus Fragilariopsis: a case study on using slide scanning, multiexpert image annotation, and image analysis in taxonomy1.

Semiautomated methods for microscopic image acquisition, image analysis, and taxonomic identification have repeatedly received attention in diatom analysis. Less well studied is the question whether and how such methods might prove useful for clarifying the delimitation of species that are difficult to separate for human taxonomists. To try to answer this question, three very similar Fragilariopsis species endemic to the Southern Ocean were targeted in this study: F. obliquecostata, F. ritscheri, and F. sublinearis. A set of 501 extended focus depth specimen images were obtained using a standardized, semiautomated microscopic procedure. Twelve diatomists independently identified these specimen images in order to reconcile taxonomic opinions and agree upon a taxonomic gold standard. Using image analyses, we then extracted morphometric features representing taxonomic characters of the target taxa. The discriminating ability of individual morphometric features was tested visually and statistically, and multivariate classification experiments were performed to test the agreement of the quantitatively defined taxa assignments with expert consensus opinion. Beyond an updated differential diagnosis of the studied taxa, our study also shows that automated imaging and image analysis procedures for diatoms are coming close to reaching a broad applicability for routine use.

SHERPA: an image segmentation and outline feature extraction tool for diatoms and other objects.

BACKGROUND: Light microscopic analysis of diatom frustules is widely used both in basic and applied research, notably taxonomy, morphometrics, water quality monitoring and paleo-environmental studies. In these applications, usually large numbers of frustules need to be identified and/or measured. Although there is a need for automation in these applications, and image processing and analysis methods supporting these tasks have previously been developed, they did not become widespread in diatom analysis. While methodological reports for a wide variety of methods for image segmentation, diatom identification and feature extraction are available, no single implementation combining a subset of these into a readily applicable workflow accessible to diatomists exists. RESULTS: The newly developed tool SHERPA offers a versatile image processing workflow focused on the identification and measurement of object outlines, handling all steps from image segmentation over object identification to feature extraction, and providing interactive functions for reviewing and revising results. Special attention was given to ease of use, applicability to a broad range of data and problems, and supporting high throughput analyses with minimal manual intervention. CONCLUSIONS: Tested with several diatom datasets from different sources and of various compositions, SHERPA proved its ability to successfully analyze large amounts of diatom micrographs depicting a broad range of species. SHERPA is unique in combining the following features: application of multiple segmentation methods and selection of the one giving the best result for each individual object; identification of shapes of interest based on outline matching against a template library; quality scoring and ranking of resulting outlines supporting quick quality checking; extraction of a wide range of outline shape descriptors widely used in diatom studies and elsewhere; minimizing the need for, but enabling manual quality control and corrections. Although primarily developed for analyzing images of diatom valves originating from automated microscopy, SHERPA can also be useful for other object detection, segmentation and outline-based identification problems.

FeatureScan: revealing property-dependent similarity of nucleotide sequences.

FeatureScan is a software package aiming to reveal novel types of DNA sequence similarity by comparing physico-chemical properties. Thirty-eight different parameters of DNA double strands such as charge, melting enthalpy, conformational parameters and the like are provided. As input FeatureScan requires two sequences, a pattern sequence and a target sequence, search conditions are set by selecting a specific DNA parameter and a threshold value. Search results are displayed in FASTA format and directly linked to external genome databases/browsers (ENSEMBL, NCBI, UCSC). An Internet version of FeatureScan is accessible at http://genome.gbf.de/featurescan/. As part of the HOBIT initiative (http://hobit.sourceforge.net/) FeatureScan is also accessible as a web service at its above home page. Currently, several preloaded genomes are provided at this Internet website (Homo sapiens, Mus musculus, Rattus norvegicus and four strains of Escherichia coli) as target sequences. Standalone executables of FeatureScan are available on request.

Culture of human neurospheres in 3D scaffolds for developmental neurotoxicity testing.

Human neural progenitor cells cultured as neurospheres are a promising tool for developmental neurotoxicity testing in vitro. In order to obtain a human cell-based tissue culture system as close to the organ as possible, it is desirable to improve the spatial organization of the "Neurosphere Assay" and use 3D scaffolds to better mimic the in vivo three dimensional cell microenvironment. For this reason we have established the conditions for short-term culture (up to 6 days) in matrigel or in IKVAV-3 peptide-functionalized hydrogels, and for long-term culture (>25 days) in IKVAV-3 peptide-functionalized hydrogels showing that these conditions support human neural progenitor cells' migration, differentiation to neurons and formation of neuronal networks. Moreover, we assessed if neurospheres grown in 3D scaffolds allow for developmental neurotoxicity compound testing. At concentrations not affecting cell viability the known developmental neurotoxic compound MeHgCl inhibits migration of human neural progenitor cells grown in 3D scaffolds with a higher potency than when the same cells are cultured on a laminin-coated surface as secondary 3D structures. Thus, this work opens the door to functional assessment of compound effects on short- and long-term cultured human neurospheres embedded in 3D scaffolds for developmental neurotoxicity testing.

PROCESS FLOW FOR CLASSIFICATION AND CLUSTERING OF FRUIT FLY GENE EXPRESSION PATTERNS.

The rapidly growing collection of fruit fly embryo images makes automated Image Segmentation and classification an indispensable requirement for a large-scale analysis of in situ hybridization (ISH) - gene expression patterns (GEP). We present here such an automated process flow for Segmenting, Classification, and Clustering large-scale sets of Drosophila melanogaster GEP that is capable of dealing with most of the complications implicated in the images.

Signal-theoretical DNA similarity measure revealing unexpected similarities of E. coli promoters.

We present an implementation of the signal theory based approach for detection of novel types of DNA similarity which are based on physical properties of DNA. Systematic study of the sensitivity of the new similarity measure revealed qualitative differences to letter-based similarity. A variety of physical parameters of DNA double strands, which in a straightforward way reflect different kinds of information hidden behind the primary structure of DNA, showed a wide range of recognition power of the signal similarity measure. We applied the novel DNA similarity measure for the analysis of promoters of E.coli genes. We found that promoter similarities revealed by our approach correlate with their transcription regulatory responsivenesses to different antibiotic and osmotic treatments. Accelerated by special hardware for fast Fourier transformations, the method is easily applicable for the analysis of entire eukaryotic genomes in minutes.

Applying signal theory to the analysis of biomolecules.

MOTIVATION: The accumulation of sequence-related and other biological data for basic research and application purposes invites disaster. It appears very likely that neither traditional thinking nor current technologies (including their foreseeable evolutionary developments) will be able to cope with this ever intensifying situation. RESULTS: We present the detailed theoretical background for applying signal theory, as known from speech recognition and image analysis, to the analysis of biomolecules. The general scheme is as follows: biochemical and biophysical properties of biomolecules are used to model an n-dimensional signal which represents the entire information-bearing biomolecule. Such signals are used to search for biological principles, analogies or similarities between biomolecules. In a series of simple experiments (bacterial DNA, generation of real signals using melting enthalpies, detection filtering by convolution of signals) we have shown that the novel system for comparative analysis of the properties of information-bearing biomolecules works as in theory. SUPPLEMENTARY INFORMATION: http://genome.gbf.de/wavepaper.

Analysis of disturbed images.

MOTIVATION: Images in cellular and molecular biology (from microscopy, blots, biochips, etc.) are often disturbed, so that the detection and analysis of the respective relevant geometrical objects may be difficult or error-prone. The disturbances are either caused by the detector, usually a CCD camera, or by the experimental setup. Furthermore, microtechnology experiments often require simultaneous multiple-colour stainings. Therefore, the image analysis of such experiments should be colour-sensitive, and colour shadings should not only be detectable but also quantifiable. RESULTS: Here, we describe a general solution as applied to the analysis of blots and DNA chips as well as to microscopy images of tissues. We decided to use (i) a stochastic filter as used by Wiener for image segmentation as the starting point for object detection, (ii) chaincodes as described by Freeman for object description, (iii) a novel 'rolling disc algorithm' to spot the objects to be analysed and (iv) an HSI instead of an RGB colour model for colour analysis. With this combination we succeeded in performing shape detection and colour-based analysis of disturbed images. AVAILABILITY: The corresponding modules (C++) are available on request.