A new method for automatic metaphase finding adaptable to different chromosome preparations

A FORTRAN computer program, running on a Digital PDP 11-34 minicomputer, has been developed for use in conjunction with a Cambridge Quantimet 720 image analyzer for the investigation of metaphase preparations in routine cytogenetics. During a short initiation phase the program is adapted to the type of metaphase being analyzed. The program is fast and its performance is good, even at low microscopic magnifications. It has other uses in biology for all investigations and characterizations of small distinct elements widely spread within a preparation (e.g., autoradiography, bacteriology).     

Novel self-cleavage activity of Staphylokinase fusion proteins: An interesting finding and its possible applications

Staphylokinase (SAK) is reported to have a serine protease domain with no proteolytic activity unlike other plasminogen activators like tissue plasminogen activator (t-PA) and urokinase. A unique protease property of Staphylokinase was observed when SAK was expressed as a fusion protein in inducible Escherichia coli expression vectors. This finding was further investigated by cloning and expressing different SAK fusions, both native and N-terminal deletions, with fusion tags like glutathione S-transferase (GST) and signal sequence of SAK in bacterial system. While all the N-terminal SAK fusions were found to self-cleave in crude and purified preparations, the C-terminal SAK fusion was stable. The cleavage property of Staphylokinase fusion proteins, inhibited by reduced glutathione and PMSF, was independent of its thrombolytic activity and also independent on the type of host employed for its expression. The serine protease domain of the SAK gene possibly lies between 20th to 77th amino acid and serine 41 of this region appears critical for such a cleavage property.     

Scientific knowledge is possible with small-sample classification

A typical small-sample biomarker classification paper discriminates between types of pathology based on, say, 30,000 genes and a small labeled sample of less than 100 points. Some classification rule is used to design the classifier from this data, but we are given no good reason or conditions under which this algorithm should perform well. An error estimation rule is used to estimate the classification error on the population using the same data, but once again we are given no good reason or conditions under which this error estimator should produce a good estimate, and thus we do not know how well the classifier should be expected to perform. In fact, virtually, in all such papers the error estimate is expected to be highly inaccurate. In short, we are given no justification for any claims.Given the ubiquity of vacuous small-sample classification papers in the literature, one could easily conclude that scientific knowledge is impossible in small-sample settings. It is not that thousands of papers overtly claim that scientific knowledge is impossible in regard to their content; rather, it is that they utilize methods that preclude scientific knowledge. In this paper, we argue to the contrary that scientific knowledge in small-sample classification is possible provided there is sufficient prior knowledge. A natural way to proceed, discussed herein, is via a paradigm for pattern recognition in which we incorporate prior knowledge in the whole classification procedure (classifier design and error estimation), optimize each step of the procedure given available information, and obtain theoretical measures of performance for both classifiers and error estimators, the latter being the critical epistemological issue. In sum, we can achieve scientific validation for a proposed small-sample classifier and its error estimate.     

Identifying genes that contribute most to good classification in microarrays

Background  

The goal of most microarray studies is either the identification of genes that are most differentially expressed or the creation of a good classification rule. The disadvantage of the former is that it ignores the importance of gene interactions; the disadvantage of the latter is that it often does not provide a sufficient focus for further investigation because many genes may be included by chance. Our strategy is to search for classification rules that perform well with few genes and, if they are found, identify genes that occur relatively frequently under multiple random validation (random splits into training and test samples).     

A combinatorial optimization approach for diverse motif finding applications

Background  

Discovering approximately repeated patterns, or motifs, in biological sequences is an important and widely-studied problem in computational molecular biology. Most frequently, motif finding applications arise when identifying shared regulatory signals within DNA sequences or shared functional and structural elements within protein sequences. Due to the diversity of contexts in which motif finding is applied, several variations of the problem are commonly studied.     

Selecting dissimilar genes for multi-class classification,an application in cancer subtyping

Background  

Gene expression microarray is a powerful technology for genetic profiling diseases and their associated treatments. Such a process involves a key step of biomarker identification, which are expected to be closely related to the disease. A most important task of these identified genes is that they can be used to construct a classifier which can effectively diagnose disease and even recognize the disease subtypes. Binary classification, for example, diseased or healthy, in microarray data analysis has been successful, while multi-class classification, such as cancer subtyping, remains challenging.     

Directional features for automatic tumor classification of mammogram images

One way for breast cancer diagnosis is provided by taking radiographic (X-ray) images (termed mammograms) for suspect patients, images further used by physicians to identify potential abnormal areas thorough visual inspection. When digital mammograms are available, computer-aided based diagnostic may help the physician in having a more accurate decision. This implies automatic abnormal areas detection using segmentation, followed by tumor classification. This work aims at describing an approach to deal with the classification of digital mammograms. Patches around tumors are manually extracted to segment the abnormal areas from the remaining of the image, considered as background. The mammogram images are filtered using Gabor wavelets and directional features are extracted at different orientation and frequencies. Principal Component Analysis is employed to reduce the dimension of filtered and unfiltered high-dimensional data. Proximal Support Vector Machines are used to final classify the data. Superior mammogram image classification performance is attained when Gabor features are extracted instead of using original mammogram images. The robustness of Gabor features for digital mammogram images distorted by Poisson noise with different intensity levels is also addressed.     

A statistical test for classification, with applications to the characterization of pathogens according to antibiotic susceptibility patterns

If n different analytical procedures are applied to a culture of a pathogen, then that pathogen is characterized by an ordered sequence or vector of length n. A group of such sequences (or 'susceptibility patterns' as they are called in this context) can be generated by applying this process to cultures taken from infected individuals in an epidemic. This paper introduced statistical tests that can be used to determine whether a pattern obtained from the culture of a single case differs from those patterns obtained from cultures of a group of infected individuals. The methods are applied to come nosocomial epidemics. These taxonomic techniques extend beyond the present application to other classification problems in the biological and medical sciences.     

High-throughput classification of images of cells transfected with cDNA clones

The sequence of the human genome has been determined. The next task is to determine the function of the genes. Classifying cellular forms of proteins encoded by human cDNA clones is a primary step toward understanding the biological role of proteins and their coding genes. We report here our ongoing work on an automatic system to facilitate this classification. Our system handles the transfection, incubation, acquisition of microscopic images of the cells, and the classification of forms there appearing in the images. Our system correctly classified proteins by their forms at a rate of 90% in feasibility studies.     

A possible association of Y chromosome heterochromatin with stature

Summary A correlation between Y chromosome length and stature was statistically analyzed in a normal male population of 142 Japanese students with a mean age of 24.0 years. Evidence was obtained that increased length of the heterochromatic band Yq12 may be associated with increased height: The correlation coefficient between band Yq12 length and height was 0.17, statistically significant at the 5% level. And, taller males had longer Y chromosomes, in which the mean length of band Yq12 was significantly longer than that of shorter males. No correlation was seen between length of the euchromatic band Yq11 and stature. The present study reveals a possible effect of Yq heterochromatin on the development of body height in man.     

End-to-end chromosome attachments in mitotic interphase and their possible significance to meiotic chromosome pairing

Cytological studies on telophase and early prophase in roottip cells of several plant species (Allium cepa, 2n=16; four Crepis species, including Crepis capillaris, 2n=6; Callitriche hermaphroditica, 2n=6; Nigella arvensis, 2n=12; Secale cereale, 2n=14) revealed that chromosome ends are attached two by two forming chains of chromosomes (interphase associations). In these chains homologous chromosomes are presumably located adjacent to each other. In Crepis capillaris it was observed that the two nucleolar chromosomes form a separate ring one end attached to the ring of the four remaining chromosomes and the other end attached to the nucleolus. It is proposed that these end-to-end attachments have significance for chromosome pairing in meiosis. The adjacent location of homologous chromosomes in the interphase associations would facilitate rapid and regular synapsis.     

A two-locus model for polymorphism for sex-linked meiotic drive modifiers with possible applications to Aedes aegypti

A two-locus model is presented which shows the possibility of maintaining a polymorphism for modifiers of sex-linked meiotic drive in the absence of fitness differences. The model is very similar to the situation actually found in some laboratory strains of the mosquito Aedes aegypti. The existence of a stable polymorphism usually requires sufficiently loose linkage between the two loci.     

On fully automatic feature measurement for banded chromosome classification

Procedures for fully automatic location of chromosome axis and centromere in metaphase chromosomes are described for a practical interactive chromosome analysis system that omits the usual stages of interactive axis and centromere correction. Accuracy of centromere finding and consequential determination of a chromosome's polarity, i.e., which end is which, is measured experimentally. The saving in interaction by not correcting centromeres is compared to the increase in errors at the classification stage and the consequent increase in interaction needed to correct these errors. Some previously unreported features for banded chromosome classification are described, and in particular a set of global shape features is introduced. The discrimination capability of the feature measurements is evaluated by use of simple statistics and by reference to the performance of classifiers trained with various feature subsets. Class discrimination capability of the global shape feature set is shown to be comparable to that of centromere position, a widely used local shape feature. The variability of feature measurements that might occur in data from different laboratories on account of differing tissue, preparation methods, and digitiser hardware is assessed using three data bases of G-banded human metaphase cells. It is shown that the differences can be considerable and that appropriate feature selection and classifier training substantially improve classification performance.     

Enhanced CellClassifier: a multi-class classification tool for microscopy images

Background  

Light microscopy is of central importance in cell biology. The recent introduction of automated high content screening has expanded this technology towards automation of experiments and performing large scale perturbation assays. Nevertheless, evaluation of microscopy data continues to be a bottleneck in many projects. Currently, among open source software, CellProfiler and its extension Analyst are widely used in automated image processing. Even though revolutionizing image analysis in current biology, some routine and many advanced tasks are either not supported or require programming skills of the researcher. This represents a significant obstacle in many biology laboratories.     

A Bayesian method for classification of images from electron micrographs

Particle classification is an important component of multivariate statistical analysis methods that has been used extensively to extract information from electron micrographs of single particles. Here we describe a new Bayesian Gibbs sampling algorithm for the classification of such images. This algorithm, which is applied after dimension reduction by correspondence analysis or by principal components analysis, dynamically learns the parameters of the multivariate Gaussian distributions that characterize each class. These distributions describe tilted ellipsoidal clusters that adaptively adjust shape to capture differences in the variances of factors and the correlations of factors within classes. A novel Bayesian procedure to objectively select factors for inclusion in the classification models is a component of this procedure. A comparison of this algorithm with hierarchical ascendant classification of simulated data sets shows improved classification over a broad range of signal-to-noise ratios.