Search

The infrequency of translocations in myelodysplastic syndromes (MDS) and chronic myelomonocytic leukemias (CMML) makes their identification and reporting interesting for the recognition of the recurrent ones and the genes involved in these neoplasias. The aims of this study were to identify new translocations associated with MDS and CMML and to establish their frequency in a cohort of 8,016 patients from the Spanish Group of MDS database. The karyotype was evaluable in 5,654 (70%) patients. Among those, 2,014 (36%) had chromosomal abnormalities, including 213 (10%) translocations identified in 195 patients. The translocations were balanced in 183 (86%) cases and unbalanced in 30 (14%) cases. All chromosomes were found to be involved in translocations, with the single exception of the Y chromosome. The chromosomes most frequently involved were in decreasing frequency: 3, 1, 7, 2, 11, 5, 12, 6, and 17. Translocations were found in karyotypes as the unique chromosomal abnormality (33%), associated with another chromosomal abnormality (11%), as a part of a complex karyotype (17%), and as a part of a monosomal karyotype (38%). There were 155 translocations not previously described in MDS or CMML and nine of them appeared to be recurrent., (Copyright © 2013 Wiley Periodicals, Inc.)

The manual analysis of the karyogram is a complex and time-consuming operation, as it requires meticulous attention to details and well-trained personnel. Routine Q-band laboratory images show chromosomes that are randomly rotated, blurred or corrupted by overlapping and dye stains. We address here the problem of robust automatic classification, which is still an open issue. The proposed method starts with an improved estimation of the chromosome medial axis, along which an established set of features is then extracted. The following novel polarization stage estimates the chromosome orientation and makes this feature set independent on the reading direction along the axis. Feature rescaling and normalizing techniques take full advantage of the results of the polarization step, reducing the intra-class and increasing the inter-class variances. After a standard neural network based classification, a novel class reassignment algorithm is employed to maximize the probability of correct classification, by exploiting the constrained composition of the human karyotype. An average 94% of correct classification was achieved by the proposed method on 5474 chromosomes, whose images were acquired during laboratory routine and comprise karyotypes belonging to slightly different prometaphase stages. In order to provide the scientific community with a public dataset, all the data we used are publicly available for download., (Copyright © 2011 Elsevier Ireland Ltd. All rights reserved.)

Multichannel chromosome image acquisition is used for cancer diagnosis and research on genetic disorders. This type of imaging, apart from aiding the cytogeneticist in several ways, facilitates the visual detection of chromosome abnormalities. However, chromosome misclassification errors result from different factors, such as uneven hybridization, spectral overlap among fluors, and biochemical noise. In this paper, we enhance the chromosome classification accuracy by making use of a region Bayes classifier that increases the classification accuracy when compared to the already developed pixel-by-pixel classifier and by incorporating the vector median filtering approach for filtering of the image. The method is evaluated using a publicly available database that contains 183 six-channel chromosome sets of images. The overall improvement on the chromosome classification accuracy is 9.99%, compared to the pixel-by-pixel classifier without filtering. This improvement in the chromosome classification accuracy would allow subtle deoxyribonucleic acid abnormalities to be identified easily. The efficiency of the method might further improve by using features extracted from each region and a more sophisticated classifier.

We developed and tested a new automated chromosome karyotyping scheme using a two-layer classification platform. Our hypothesis is that by selecting most effective feature sets and adaptively optimizing classifiers for the different groups of chromosomes with similar image characteristics, we can reduce the complexity of automated karyotyping scheme and improve its performance and robustness. For this purpose, we assembled an image database involving 6900 chromosomes and implemented a genetic algorithm to optimize the topology of multi-feature based artificial neural networks (ANN). In the first layer of the scheme, a single ANN was employed to classify 24 chromosomes into seven classes. In the second layer, seven ANNs were adaptively optimized for seven classes to identify individual chromosomes. The scheme was optimized and evaluated using a "training-testing-validation" method. In the first layer, the classification accuracy for the validation dataset was 92.9%. In the second layer, classification accuracy of seven ANNs ranged from 67.5% to 97.5%, in which six ANNs achieved accuracy above 93.7% and only one had lessened performance. The maximum difference of classification accuracy between the testing and validation datasets is <1.7%. The study demonstrates that this new scheme achieves higher and robust performance in classifying chromosomes.

The method of spectral karyotyping (SKY) is based on a combination of the technologies of charge-coupled device imaging and spectrometry. The engineering feasibility has been realized in the SpectraCube system from Applied Spectral Imaging Inc., and it allows the simultaneous identification of all 24 human chromosomes. This is performed by characterizing the spectral signature of every image pixel in relation to a fluorochrome combinatorial library translating the image and spectral information into chromosome classification. Applications for SKY include pre- and postnatal characterization of certain numerical and structural rearrangements and complex karyotypes and highly informative analysis of sample materials with only single or few cells available for investigation.

Cytogenetics is a key tool in the detection of acquired chromosomal abnormalities and in the diagnosis of genetic diseases such as leukemia. The karyotyping is a set of procedures, in the scope of the cytogenetics, that produces a visual representation of the 46 chromosomes (called karyogram), paired and arranged in decreasing order of size. The pairing procedure aims to identify all pairs of homologous chromosomes.The pairing criterion is based on dimensional, morphological,and textural features similarity. This process is time consuming when performed manually, and demanding from a technical point of view. An automatic pairing algorithm would thus bring benefits, but it remains an open problem to date.In this paper a new strategy for automatic pairing of homologous chromosomes is proposed. Besides the traditional features described in the literature, the Mutual Information (MI) is used to discriminate chromosome textural differences. A supervised non-linear classifier is trained by using manual classifications provided by expert technicians, combining the different features computed from each pair.Simulations using 836 real chromosome images, obtained with a Leica Optical Microscope DM 2500, in a leave-one-out cross validation fashion, were performed for training and testing the algorithm.Promising and relevant results were found, despite the poor quality of the original chromosome images, contrasting with state-of the-art algorithms and datasets found in the literature.

The manual analysis of the karyogram is a complex, wearing and time-consuming operation. It requires a very meticulous attention to details and calls for well-trained personnel. Even though existing commercial software packages provide a reasonable support to cytogenetists, they very often require human intervention to correct challenging situations. We developed a robust automatic classification system conceived to cope with routine images in which chromosomes are randomly rotated, possibly blurred or also corrupted by overlapping or by dye stains. It consists in a sequence of modules comprising robust feature extraction based on medial axis, chromosome polarization, feature pre-processing, and Neural Network classification followed by a class reassigning algorithm.We show the effectiveness of the proposed method on data comprising karyotypes belonging to slightly different stage of the prometaphase. This dataset contains 119 karyotypes (5474 chromosomes), 70 of which were used for training and validation and 49 for the final testing. In this latter set of images, the system achieved a classification accuracy, as compared to manual ground truth, of 95.6%.

To identify genetic events underlying the genesis and progression of multiple myeloma (MM), we conducted a high-resolution analysis of recurrent copy number alterations (CNAs) and expression profiles in a collection of MM cell lines and outcome-annotated clinical specimens. Attesting to the molecular heterogeneity of MM, unsupervised classification using nonnegative matrix factorization (NMF) designed for array comparative genomic hybridization (aCGH) analysis uncovered distinct genomic subtypes. Additionally, we defined 87 discrete minimal common regions (MCRs) within recurrent and highly focal CNAs. Further integration with expression data generated a refined list of MM gene candidates residing within these MCRs, thereby providing a genomic framework for dissection of disease pathogenesis, improved clinical management, and initiation of targeted drug discovery for specific MM patients.

M-FISH (multicolor fluorescence in situ hybridization) is a recently developed cytogenetic technique for cancer diagnosis and research on genetic disorders which uses 5 fluors to label uniquely each chromosome and a fluorescent DNA stain. In this paper, an automated method for chromosome classification in M-FISH images is presented. The chromosome image is initially decomposed into a set of primitive homogeneous regions through the morphological watershed transform applied to the image intensity gradient magnitude. Each segmented area is then classified using a Bayes classifier. We have evaluated our methodology on a commercial available M-FISH database. The classifier was trained and tested on non-overlapping chromosome images and an overall accuracy of 89% is achieved. By introducing feature averaging on watershed basins, the proposed technique achieves substantially better results than previous methods at a lower computational cost.

Since the birth of chromosome analysis by the aid of computers, building a fully automated chromosome analysis system has been the ultimate goal. Along with many other challenges, automating chromosome classification and segmentation has been one of the major challenges especially due to overlapping and touching chromosomes. In this paper we present a novel decomposition method for overlapping and touching chromosomes in M-FISH images. To overcome the limited success of previous decomposition methods that use partial information about a chromosome cluster, we have incorporated more knowledge about the clusters into a maximum-likelihood frame work. The proposed method evaluates multiple hypotheses based on geometric information, pixel classification results, and chromosome sizes, and a hypothesis that has a maximum-likelihood is chosen as the best decomposition of a given cluster. About 90% of accuracy was obtained for two or three chromosome clusters, which consist about 95% of all clusters with two or more chromosomes.

Traditional chromosome imaging has been limited to grayscale images, but recently a 5-fluorophore combinatorial labeling technique (M-FISH) was developed wherein each class of chromosomes binds with a different combination of fluorophores. This results in a multispectral image, where each class of chromosomes has distinct spectral components. In this paper, we develop new methods for automatic chromosome identification by exploiting the multispectral information in M-FISH chromosome images and by jointly performing chromosome segmentation and classification. We (1) develop a maximum-likelihood hypothesis test that uses multispectral information, together with conventional criteria, to select the best segmentation possibility; (2) use this likelihood function to combine chromosome segmentation and classification into a robust chromosome identification system; and (3) show that the proposed likelihood function can also be used as a reliable indicator of errors in segmentation, errors in classification, and chromosome anomalies, which can be indicators of radiation damage, cancer, and a wide variety of inherited diseases. We show that the proposed multispectral joint segmentation-classification method outperforms past grayscale segmentation methods when decomposing touching chromosomes. We also show that it outperforms past M-FISH classification techniques that do not use segmentation information.

Chromosomes are essential genomic information carriers. Chromosome classification constitutes an important part of routine clinical and cancer cytogenetics analysis. Cytogeneticists perform visual interpretation of banded chromosome images according to the diagrammatic models of various chromosome types known as the ideograms, which mimic artists' depiction of the chromosomes. In this paper, we present a subspace-based approach for automated prototyping and classification of chromosome images. We show that 1) prototype chromosome images can be quantitatively synthesized from a subspace to objectively represent the chromosome images of a given type or population, and 2) the transformation coefficients (or projected coordinate values of sample chromosomes) in the subspace can be utilized as the extracted feature measurements for classification purposes. We examine in particular the formation of three well-known subspaces, namely the ones derived from principal component analysis (PCA), Fisher's linear discriminant analysis, and the discrete cosine transform (DCT). These subspaces are implemented and evaluated for prototyping two-dimensional (2-D) images and for classification of both 2-D images and one-dimensional profiles of chromosomes. Experimental results show that previously unseen prototype chromosome images of high visual quality can be synthesized using the proposed subspace-based method, and that PCA and the DCT significantly outperform the well-known benchmark technique of weighted density distribution functions in classifying 2-D chromosome images.

Background: Allelic-loss studies record data on the loss of genetic material in tumor tissue relative to normal tissue at various loci along the genome. As the deletion of a tumor suppressor gene can lead to tumor development, one objective of these studies is to determine which, if any, chromosome arms harbor tumor suppressor genes., Results: We propose a large class of mixture models for describing the data, and we suggest using Bayes factors to select a reasonable model from the class in order to classify the chromosome arms. Bayes factors are especially useful in the case of testing that the number of components in a mixture model is n0 versus n1. In these cases, frequentist test statistics based on the likelihood ratio statistic have unknown distributions and are therefore not applicable. Our simulation study shows that Bayes factors favor the right model most of the time when tumor suppressor genes are present. When no tumor suppressor genes are present and background allelic-loss varies, the Bayes factors are often inconclusive, although this results in a markedly reduced false-positive rate compared to that of standard frequentist approaches. Application of our methods to three data sets of esophageal adenocarcinomas yields interesting differences from those results previously published., Conclusions: Our results indicate that Bayes factors are useful for analyzing allelic-loss data.

Purpose: To investigate whether unstable types of chromosomal aberrations are more effective in priming apoptotic cell death in comparison with stable ones. Also, to highlight the phase of the cell cycle at which apoptosis occurs and the mechanism of its execution., Materials and Methods: G0 human peripheral blood lymphocytes were X-irradiated in the presence or absence of the repair inhibitor cytosine arabinoside (Ara-C). After irradiation, the lymphocytes were analysed for induction of dicentrics, translocations, apoptosis, p53 and survivin expression at various recovery times., Results: A preferential elimination of cells bearing dicentrics with respect to those with balanced translocations was observed. There was a time-dependent correlation between the decrease in the frequency of dicentrics and the increase in the per cent of apoptotic cells. Most of the apoptotic cells were labelled with bromodeoxyuridine and were mononucleated in cytochalasin B-treated cells cultures (blocked cytokinesis). However, after continuous colcemid treatment, the apoptotic pathway was not induced. Moreover, in the G2/M-phase, an increase in p53 and a decrease in survivin occurred that were X-ray and Ara-C dose dependent., Conclusions: The apoptotic process is primed when the dicentric-bearing human peripheral blood lymphocytes attempt to exit from metaphase. It is possible that unstable aberrations generate changes in the mitotic spindle causing mechanical tension at the kinetochore, activating the mitotic checkpoint and the execution of p53/survivin-dependent apoptosis.

Background: Comparative genomic hybridization (CGH) is a relatively new molecular cytogenetic method for detecting chromosomal imbalance. Karyotyping of human metaphases is an important step to assign each chromosome to one of 23 or 24 classes (22 autosomes and two sex chromosomes). Automatic karyotyping in CGH analysis is needed. However, conventional karyotyping approaches based on DAPI images require complex image enhancement procedures., Methods: This paper proposes a simple feature extraction method, one that generates density profiles from original true color CGH images and uses normalized profiles as feature vectors without quantization. A classifier is developed by using support vector machine (SVM). It has good generalization ability and needs only limited training samples., Results: Experiment results show that the feature extraction method of using color information in CGH images can improve greatly the classification success rate. The SVM classifier is able to acquire knowledge about human chromosomes from relatively few samples and has good generalization ability. A success rate of moe than 90% has been achieved and the time for training and testing is very short., Conclusions: The feature extraction method proposed here and the SVM-based classifier offer a promising computerized intelligent system for automatic karyotyping of CGH human chromosomes., (Copyright 2001 Wiley-Liss, Inc.)

Two digital fluorescence microscopy systems, spectral karyotyping (SKY) and multiplex fluorescence in situ hybridisation (M-FISH), are used with multicolour probe sets to assist in the detection of chromosome aberrations. We have compared the resolution of the two methods in their ability to identify karyotype rearrangements, which have occurred during the divergence of Homo sapiens and Hylobates leucogenys in evolution. A 24-color human paint kit distinguishes 74 conserved autosomal segments in H. leucogenys, some of which are difficult to resolve. We examined the extent to which the SKY and M-FISH techniques are able to detect the smallest of these bands. We have found this to be a rigorous test of multicolour chromosome classification systems. We conclude from our results that both systems are able invariably to classify the majority of conserved segments but differ in the efficiency of detection of small inserts., (Copyright 2001 Wiley-Liss, Inc.)

The metaphase nucleolar organizer regions (NORs) contain ribosomal genes associated with proteins such as upstream binding factor (UBF) and RNA polymerase I (RPI). These genes are clustered in 10 loci of the human acrocentric chromosomes (13, 14, 15, 21, and 22). Some NOR-associated proteins, termed AgNOR proteins, can be specifically stained by silver. In this study we took advantage of technical advances in digital imaging, image restoration techniques, and factorial correspondence analysis (FCA) to study the different AgNOR staining patterns of metaphase chromosomes in human lymphocytes. Three predominant patterns could be distinguished: pair (47%), stick-like (28%), and unstained (18%) structures. By studying the frequency of occurrence of each pattern on different chromosomes, two groups could be defined. Chromosomes 13, 14, and 21 carried predominantly pair or stick-like AgNOR structures, whereas chromosomes 15 and 22 mainly carried pair AgNOR structures or remained unstained. We suggest that the different AgNOR shapes reflect both the number of ribosomal genes carried by each chromosome and the differential recruitment of active ribosomal genes in each NOR cluster. This is the first study showing a nonrandom distribution of AgNOR shape among acrocentric chromosomes.

A colored banding pattern for human chromosomes is described that distinguishes each chromosome in a single fluorescence in situ hybridization with a set of subregional DNA probes. Alu/polymerase chain reaction products of various human/rodent somatic cell hybrids (fragment hybrids) were pooled into two probe sets that were labeled differentially and detected by red and green fluorescence. Chromosome regions hybridized by DNA present in both pools appeared yellow. The result was a multi-color set of 110 distinct signals per haploid chromosome set for the human karyotype. Each individual chromosome showed a unique sequence of signals, a result termed the "chromosome bar code". The reproducibility of the hybridization pattern in various labeling and hybridization experiments was analyzed by computer densitometry. We have applied the chromosome bar code both in diagnostic cytogenetics and in genome studies. The approach allows the rapid identification of chromosomes and chromosome rearrangements. Although not yet showing the resolution of classical banding patterns, the present experiments demonstrate various applications in which the present multi-color bar code can significantly add to the spectrum of cytogenetic techniques.

A multilayer perceptron (MLP) neural network (NN) has been studied for human chromosome classification. Only 10-20 examples were required for the MLP NN to reach its ultimate performance classifying chromosomes of 5 types. The empirical dependence of the entropic error on the number of examples was found to be highly comparable to the 1/t function. The principal component analysis (PCA) was used, both for network initialization and for feature reduction purposes. The PCA demonstrated the importance of retaining most of the image information whenever small training sets are used. The MLP NN classifier outperformed the Bayes piecewise classifier for all the cases tested. The MLP classifier was found to be almost unsusceptible to the ratio of the number of training vectors to the number of features, whereas the piecewise classifier was highly dependent on this ratio.

Automated chromosome classification from metaphase spreads has been a difficult research problem over the past 30 years. Numerous techniques have been implemented to address the research problem. Image processing techniques support many methods utilized for automated and/or semiautomated karyotyping. Some current systems correctly classify individual chromosomes at high rates but lack the capability to properly classify chromosomes for entire cells consistently. Most systems attain high classification rates ignoring overlapping metaphase chromosomes for testing purposes. Chromosome classification depends on identifying features common to each chromosome class or number. Some of the common features incorporated into various chromosome classification systems include length, centromere location, banding pattern, and width. To complicate classification, chromosome features tend to vary not only between people but from cell to cell for the same person. Additionally, chromosomes found in metaphase spreads may have any orientation and virtually any degree of overlap with other chromosomes present. Besides the inherent barriers impeding chromosome classification, many systems perform image processing operations to the chromosomes for feature determination. Image processing operations, such as image rotation, that manipulate the chromosome grey-level information may distort the feature calculations utilized in karyotyping. Consequently, feature stability becomes an important issue for improving karyotyping capability.

This paper describes the application of a probabilistic neural network (PNN) to the classification of normal human chromosomes. The inputs to the network are 30 different features extracted from each chromosome in digitized images of metaphase spreads. The output is 1 of 24 different classes of chromosomes (the 22 autosomes plus the sex chromosomes X and Y). An updating procedure was implemented to take advantage of the fact that in a normal somatic cell only two chromosomes can be assigned to each class. The network has been tested using the Copenhagen, Edinburgh, and Philadelphia databases of digitized images of human chromosomes. The recognition rates achieved in this study are superior to those reported using either the maximum likelihood or back propagation neural network techniques.

Human chromosomes, represented by band transition sequences, chromosome area, centromeric index by area and centromeric index by density, were karyotyped by computer. A reference set of chromosomes provided frequencies of occurrence of each density class and difference class of the band transition sequence as well as of each of the three global features. The karyotyping program was designed to handle all metaphases, even those from which severely bent and overlapped chromosomes were excluded. In one experiment, 21 metaphases were karyotyped on the basis of a reference set and the results were compared with earlier results of visual analysis of band transition profiles developed from band transition sequences: 0.8% errors were made in the visual experiment and 1.4% errors were made in the computer based experiment. In a second experiment, 179 metaphases were divided into reference and test sets and karyotyped by computer with an error rate of 3.4%. By further analysis it was found that metaphases with many misclassified chromosomes could often be automatically distinguished from metaphases with few errors. Thus by automatic rejection of 7% of the metaphases the error rate could be reduced to 2.6%. The computer program for chromosome karyotyping will now be implemented in a semi-automatic system for practical clinical chromosome analysis.

Chromosomes were isolated from Chinese hamster x human hybrid cell lines containing four and nine human chromosomes. Human genomic DNA was biotinylated by nick translation and used to label the human chromosomes by in situ hybridization in suspension. Streptavidin was covalently coupled to the surface of magnetic beads and these were incubated with the hybridized chromosomes. The human chromosomes were bound to the magnetic beads through the strong biotin-streptavidin complex and then rapidly separated from nonlabeled Chinese hamster chromosomes by a simple permanent magnet. The hybridization was visualized by additional binding of avidin-FITC (fluorescein) to the unoccupied biotinylated human DNA bound to the human chromosomes. After magnetic separation, up to 98% of the individual chromosomes attached to magnetic beads were classified as human chromosomes by fluorescence microscopy.

The behavior of individual allocyclic chromosomes has been analyzed in lymphocytes of a sister and a brother with Bloom's syndrome. Of 4,633 46 diploid cells, 115 showed allocyclic chromosomes, and 74 of these had 44, 45 or 46 normal metaphase chromosomes accompanied by one or two allocyclic chromosomes. Of 56 tetraploid cells, 9 contained such chromosomes. The allocyclic chromosomes appeared "pulverized" or extended corresponding to S or G2 PCC. We have proposed the hypothesis that individual allocyclic chromosomes do not, as a rule, come from micronuclei, as has often been assumed, but have been left behind in their cycle. This would be caused by a mutation or deletion of a hypothetical coiling center situated near the centromere of each chromosome arm. The following observations agree with our explanation but less well or not at all with the idea of micronuclei: (1) In only 9.6% of the cells does the allocyclic chromosome lie at the edge of the metaphase plate. (2) In 24 cells a part of a chromosome is "pulverized" while the rest is in metaphase. (3) Both a "pulverized" and an extended chromosome were present in the same cell. (4) A "pulverized" acrocentric is often nose-to-nose with a normal D or G chromosome. (5) No allocyclic chromosomes corresponding to G1 PCC have been found in our material. (6) When a ring is replaced by an allocyclic chromosome, it is usually a member of a 46-chromosome complement. Furthermore, the occurrence of allocyclic chromosomes is correlated with that of other chromosome anomalies which do not follow a Poisson distribution. Allocyclic chromosomes are also more frequent (16%) in tetraploid than in diploid cells (2%).

The statistical analysis of the distribution of human chromosomes length served as a basis for their classification according to the suggestions of the Denver Conference (1960). Standards of metaphasal chromosome length were used for programmes of computer-assisted classification. The introduction of banded chromosomes in the study provided a new measurable feature which was used for classification standards accepted at the Paris Conference (1971). The measurements of these two features have been used also in the mathematical models of computer analysis and classification of chromosomes. In the present study the relationship between the number of bands and the length of chromosomes was studied in the period of cell division from prophase to metaphase in the cultures of leucocytes of healthy people. This relationship is strongly correlated and can be approximated in description to simple regression in the period from prophase to metaphase. The relationship will facilitate the construction of dynamic patterns for the needs of computer-assisted classification of chromosomes and detection of structural aberrations.

The basic technical facts of human cytogenetics and the laboratory methods employed in chromosome research are explained in simple terms. The main variables used to describe chromosome images are defined and discussed. Three discriminant analysis models for chromosome classification are developed: one in which each chromosome is classified in isolation, a modification in which the cell, if normal, contains 2 chromosomes of each of the 23 kinds, and a final one in which the cell is the unit of analysis instead of the chromosome. Suggestions are made to reduce the calculations involved and to take into account missing chromosomes. The problem of detection and classification of aberrative chromosomes is studied, also in relation to multiple cell analysis. Finally four relevant problems are briefly discussed: selection of metaphase spreads, selection of variables, uncertain reference classification and measurement of performance.

A procedure for automatic classification of G-banded human chromosomes has been implemented on a semiautomated system for routine clinical analysis. Chromosomes represented by their density profiles are described by so-called weighted density distributions (WDDs) by application of a number of weighting functions and classified by a parametric discriminant analysis. During 16 mo of routine use of the system, 2,794 metaphases (127,925 chromosomes) from amniotic fluid have been karyotyped by the system with an error rate of 8-9%. This corresponds to 4-5 errors per metaphase. These errors can immediately be corrected by the operator on a displayed karyogram with a light pen.