Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Feature selection”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 613 records · Page 34Linked to original sources
PubMed2005

Time-frequency analysis of target detection reveals an early interface between bottom-up and top-down processes in the gamma-band.

The early visual gamma-band response is an oscillatory signal evoked approximately 100 ms after stimulation. While some studies have found effects of various cognitive processes on this signal, such effects could not be replicated in other studies. Accordingly, some authors have claimed that evoked gamma-band activity reflects merely sensory functions. To resolve these conflicting positions, we conducted a target detection experiment in which the feature that defined the target could be distributed over a large or a small part of the entire stimulus. Only targets covering a larger area of the entire stimulus evoked stronger gamma-band activity than standards although the over-all stimulus size was identical for all stimuli. This increase in evoked activity resulted from stronger oscillatory power and not exclusively from stronger phase-locking. In contrast, N1 and P3 amplitudes were larger for target stimuli irrespective of the distribution of the relevant stimulus feature. These results are consistent with the notion that early gamma-band activity is generated by feature-selective neural assemblies the activity of which can in fact be modulated by top-down processes. This interaction, however, may be only detectable in scalp-recorded EEG if it affects a sufficient number of neural assemblies.

Adult↗
PubMedDate not supplied

High-resolution image cytometry of rat sperm nuclear shape, size and chromatin status. Experimental validation with the reproductive toxicant vinclozolin.

Recent studies have shown that the complex inter-related processes of sperm chromatin organization and nuclear morphogenesis, both of which are important fertility determinants, may be disrupted by chemicals. A high-resolution image cytometry method has been developed, using the fluorescent dye bisbenzimide, for the measurement of 20 features of the sperm nucleus related to size, form and chromatin status in the rat. For the complete set of features measured and from a total of 150 spermatozoa assessed per sample, the overall coefficient of reproducibility was 5%. Then, an experimental validation of the method was carried out in rats chronically exposed to the antiandrogenic reproductive toxicant vinclozolin and control animals. Univariate statistics revealed significant vinclozolin-induced changes for 19 out of 20 morphometric and chromatin features. Stepwise linear discriminant analysis classified correctly 84.3% of the sperm nuclei with only four features selected. The accuracy and reproducibility of the cytometry assessment of the sperm nuclei together with the results of the experimental validation suggest this method may be a new powerful tool for use in reproductive toxicology.

Androgen Antagonists↗
PubMed2026

Unveiling the power of TIIC: A prognostic tool for esophageal adenocarcinoma.

BACKGROUND: Esophageal adenocarcinoma (EAC) remains a lethal malignancy with limited prognostic tools for guiding immunotherapy. Tumor-infiltrating immune cells (TIICs) play a critical role in EAC prognosis and treatment response. METHODS: We integrated single-cell RNA sequencing and bulk transcriptome data from TCGA and GEO databases. TIIC-specific RNAs were identified via tissue specificity index calculation combined with machine learning feature selection. Twenty machine learning algorithms were benchmarked to construct an optimal TIIC signature score (TIIC-Score) based on the comprehensive C-index. Immunotherapy response, genomic mutation, and copy number variation were analyzed. Summary-data-based Mendelian randomization (SMR) and two-sample Mendelian randomization (MR) were performed to explore genetic associations. Core prognostic TIIC-related genes were functionally validated in esophageal cancer cell lines through loss-of-function assays. RESULTS: The TIIC-Score demonstrated robust prognostic value for 1-, 2-, and 3-year overall survival across multiple cohorts, outperforming 22 published models. High TIIC-Score was associated with poor survival and increased chromosomal instability. Mutation profiling revealed high frequencies of TP53 (78.2%), TTN (48.7%), and SYNE1 (30.8%). MR analysis identified a significant association between gastro-oesophageal reflux and EAC risk at SNP rs8130507. Functionally, CCNI was upregulated in esophageal cancer cells, and its knockdown suppressed malignant phenotypes while promoting apoptosis, supporting its pro-tumorigenic role. CONCLUSION: The TIIC-Score provides a novel prognostic framework for EAC that effectively stratifies patient risk and may help identify individuals most likely to benefit from immunotherapy.

Esophageal adenocarcinoma↗
PubMed2005

Envelope gene capture and insect retrovirus evolution: the relationship between errantivirus and baculovirus envelope proteins.

In this report the evolution of insect retroviruses (errantiviruses) is reviewed with particular emphasis on the relationship between their env protein and a baculovirus envelope fusion protein. In addition, selected features of the env protein from the errantivirus Dme17.6V are examined. These include characterization of the 21 amino acid predicted fusion peptide sequence that is highly homologous to a region of baculovirus envelope fusion proteins. We found that, although this sequence could not substitute for the homologous sequence in the baculovirus LD130 envelope fusion protein, by changing four amino acids, the hybrid construct became active for low-pH induced cell fusion. In addition, a Dme17.6V env-egfp construct was found to localize to cell membranes.

Amino Acid Sequence↗
PubMed2026

Development and validation of a serum peptidomic signature for early detection of asymptomatic ovarian cancer: A multi-center prospective study.

Early detection of asymptomatic ovarian cancer (asym-OC) remains a critical challenge, the failure of which underlies its high mortality. Performing serum peptidomic profiling of 843 participants in the cohort SOCFCP, we distill 1,081 initial features into a 7-marker panel for asym-OC detection via a biology-informed machine-learning (ML)-based feature selection strategy. Three markers significantly revert toward non-OC levels after surgery. Integrating the panel with age, CA125, and HE4, we develop and externally validate (n = 159) a LightGBM model, ProMS+. For early-stage OC detection, ProMS+ shows a specificity of 92.6% at 95.0% sensitivity, outperforming CA125 (44.7%), HE4 (11.2%), and Risk of Ovarian Malignancy Algorithm (ROMA) (24.0%), with an area under the curve (AUC) of 0.993. In a simulated high-risk population (n = 100,000; OC prevalence = 1%), ProMS+ yields a high AUC (0.983) and a higher positive predictive value than CA125, HE4, and Age + CA125 + HE4 combined model (0.201 vs. 0.027, 0.090, and 0.064). ProMS+ offers a promising, non-invasive, and interpretable approach for the early detection of asym-OC.

Humans↗
PubMed2006

Phenotype definition in epilepsy.

Phenotype definition consists of the use of epidemiologic, biological, molecular, or computational methods to systematically select features of a disorder that might result from distinct genetic influences. By carefully defining the target phenotype, or dividing the sample by phenotypic characteristics, we can hope to narrow the range of genes that influence risk for the trait in the study population, thereby increasing the likelihood of finding them. In this article, fundamental issues that arise in phenotyping in epilepsy and other disorders are reviewed, and factors complicating genotype-phenotype correlation are discussed. Methods of data collection, analysis, and interpretation are addressed, focusing on epidemiologic studies. With this foundation in place, the epilepsy subtypes and clinical features that appear to have a genetic basis are described, and the epidemiologic studies that have provided evidence for the heritability of these phenotypic characteristics, supporting their use in future genetic investigations, are reviewed. Finally, several molecular approaches to phenotype definition are discussed, in which the molecular defect, rather than the clinical phenotype, is used as a starting point.

Epilepsy↗
PubMed2005

Diffuse reflectance patterns in cervical spectroscopy.

OBJECTIVES: Our laboratory seeks to develop minimally invasive cost-effective methods to improve screening and detection of curable precursors to cervical cancer. Previously, we have presented pilot studies that assess the diagnostic power of auto-fluorescence and diffuse reflectance spectroscopy. In the present study, we evaluate diffuse reflectance spectra from a comprehensive 850 patient clinical trial to determine its ability to discriminate normal tissue from several grades of abnormal cervical tissue. METHODS: Diffuse reflectance spectra at four source detector separations measured from 549 cervical sites were available for analysis. Three classifiers were implemented: one used spectral data directly as input, a second used simple spectral features such as peak position and intensity, and one used principal component analysis for feature selection. Algorithms were developed and evaluated using leave-one-out cross-validation to classify normal and precancerous cervical tissue. The percentage of samples correctly classified was used to evaluate and compare the performance of the algorithms, as compared to histology. RESULTS: Diffuse reflectance spectra of cervical precancer showed consistent differences from that of normal tissue at all source detector separations; reflectance intensity of precancer was lower than that of normal tissue on average. Normal cervical tissue spectra show more intensity variation between patients than other tissue grades. Reflectance spectra acquired from the closest source detector separations consistently demonstrated the most relevant information for tissue classification. Two persistent spectral patterns demonstrated that the contribution of hemoglobin absorption and the wavelength-dependent spectral slope contained relevant information for classification. CONCLUSIONS: Spectral patterns in diffuse reflectance spectra can be used for the discrimination of normal cervical tissue from low grade and high grade squamous intraepithelial lesions.

Algorithms↗
PubMed2000

Proliferation in HHV-8-positive primary effusion lymphomas is associated with expression of HHV-8 cyclin but independent of p27(kip1).

Primary effusion lymphoma (PEL) develops in immunodeficient patients, selectively localizes to the serous body cavities, and harbors infection by human herpesvirus type-8 (HHV-8), also known as Kaposi's sarcoma-associated herpesvirus. HHV-8 encodes a viral (v)-cyclin homologous to cellular D-type cyclins, a class of positive cell-cycle regulators that are physiologically modulated by the p27(Kip1) cell cycle inhibitor. The aims of the present study were: 1) to establish the expression pattern of p27(Kip1) in PEL; and 2) to address the relationship between p27(Kip1) expression, proliferation index, and expression of cellular cyclin D1 and v-cyclin in PEL. Expression of p27(Kip1) was detected in all (n = 18) PEL samples analyzed by both immunocytochemistry and Western blot. All PELs displayed a high proliferation index as assessed by Ki-67 staining. Expression of cellular cyclin D1 was absent in all PELs tested, which conversely expressed (14 out of 14 samples) v-cyclin by immunocytochemistry and/or Western blot. In contrast to PELs, HHV-8-negative lymphomatous effusions secondary to a tissue-based lymphoma generally failed to express p27(Kip1). Overall, these data show that PELs consistently express p27(Kip1) protein despite the high proliferative rate of the lymphoma clone, suggesting that p27(Kip1) may be unable to drive cell-cycle arrest in PEL cells. The co-existence of p27(Kip1) expression and high proliferative index is a selective feature of PEL among lymphomas involving the serous body cavities, because lymphomatous effusions secondary to a tissue-based lymphoma generally display the inverse relationship between p27(Kip1) positivity and growth fraction observed in normal lymphoid tissues and in most other lymphomas. Expression of p27(Kip1) in PEL associates with expression of HHV-8 v-cyclin, but not of cellular cyclin D1. The fact that HHV-8 v-cyclin is resistant to p27(Kip1)-modulated inhibition, whereas cellular cyclin D1 is sensitive, may explain, at least in part, the co-existence of p27(Kip1) expression and high proliferative index observed in PEL.

Blotting, Western↗
PubMed2003

Bcl-2 promotes premature senescence induced by oncogenic Ras.

The expression of the apoptosis inhibitory protein, Bcl-2, is increased in naturally senescing human fibroblasts and upon induction of their senescence-like growth arrest by oxidative stress, implying its role in maintaining their extended viability. Oncogenic Ras(V12) protein induces signaling cascades that result in the premature senescence of primary fibroblast cells, which are insensitive to oncogene-dependent apoptosis. Here we show that constitutive expression of Bcl-2 accelerates selected features of the Ras-induced senescence program in primary human fibroblasts. Yet, Bcl-2 also inhibits fibroblast apoptosis induced by exogenous H(2)O(2), while both signals induce an increased endogenous Bcl-2 expression in these cells. Together, these data suggest a context-dependent phenotypic function of Bcl-2 in the regulation of overlapping cell fate specification programs, with potential implications for both physiology and multistep tumorigenesis.

Apoptosis↗
PubMed2003

Highly accurate classification of Watson-Crick basepairs on termini of single DNA molecules.

We introduce a computational method for classification of individual DNA molecules measured by an alpha-hemolysin channel detector. We show classification with better than 99% accuracy for DNA hairpin molecules that differ only in their terminal Watson-Crick basepairs. Signal classification was done in silico to establish performance metrics (i.e., where train and test data were of known type, via single-species data files). It was then performed in solution to assay real mixtures of DNA hairpins. Hidden Markov Models (HMMs) were used with Expectation/Maximization for denoising and for associating a feature vector with the ionic current blockade of the DNA molecule. Support Vector Machines (SVMs) were used as discriminators, and were the focus of off-line training. A multiclass SVM architecture was designed to place less discriminatory load on weaker discriminators, and novel SVM kernels were used to boost discrimination strength. The tuning on HMMs and SVMs enabled biophysical analysis of the captured molecule states and state transitions; structure revealed in the biophysical analysis was used for better feature selection.

Algorithms↗
PubMed1981

Diffusion coefficients of hemoglobin by intensity fluctuation spectroscopy: effects of varying pH and ionic strength.

Measurements of the mutual diffusion coefficients (D) of the liganded human hemoglobins (Hb) oxy-HbA and oxy-HbS were performed as a function of Hb concentration (CHb), pH, and ionic strength (tau) by intensity fluctuation spectroscopy (IFS). Average diffusion coefficients, (D), and normalized variances, ((D/(D) - 1)2), were recorded. Results are reported and select features are discussed quantitatively. (a) for tau = 0.15 M, the shape of the (d) vs. CHb curve is found to vary with pH. We developed a precise description of this effect in the form of an algebraic relationship between (D), CHb, and Z, the titration charge. (b) only slight differences between the (D) values of oxy-HbS and oxy-HbA are observed, at tau = 0.15 M, for CHb Less Than or Equal To 10 g%. These differences are explained by the theory of part a. (c) No evidence of aggregation is found in solutions of oxy-HbA or oxy-HbS, at tau = 0.15 M, for CHb Less Than or Equal To 10 g%. (d) Indications of aggregation appear in oxy-HbA solutions at very low concentrations of salt. An estimate is made of the extent of aggregation, and the average radius of a cluster is determined.

Diffusion↗
PubMed2004

Efficient and cost-effective experimental determination of kinetic constants and data: the success of a Bayesian systematic approach to drug transport, receptor binding, continuous culture and cell transport kinetics.

Details about the parameters of kinetic systems are crucial for progress in both medical and industrial research, including drug development, clinical diagnosis and biotechnology applications. Such details must be collected by a series of kinetic experiments and investigations. The correct design of the experiment is essential to collecting data suitable for analysis, modelling and deriving the correct information. We have developed a systematic and iterative Bayesian method and sets of rules for the design of enzyme kinetic experiments. Our method selects the optimum design to collect data suitable for accurate modelling and analysis and minimises the error in the parameters estimated. The rules select features of the design such as the substrate range and the number of measurements. We show here that this method can be directly applied to the study of other important kinetic systems, including drug transport, receptor binding, microbial culture and cell transport kinetics. It is possible to reduce the errors in the estimated parameters and, most importantly, increase the efficiency and cost-effectiveness by reducing the necessary amount of experiments and data points measured.

Bayes Theorem↗
PubMed1999

New ligand, N-(2-pyridylmethyl)aminoacetate, for use in the immobilised metal ion affinity chromatographic separation of proteins.

A new chelating compound has been developed for use in the immobilised metal ion affinity chromatographic separation of proteins. The tridentate ligand, sodium N-(2-pyridylmethyl)aminoacetate (carbpyr), 1, was prepared via a one-step synthesis from 2-picolylamine, 3 and then immobilised onto Sepharose CL-4B through the epoxide coupling procedure. The binding behaviour of the resulting IMAC sorbent, following chelation with Cu2+ ions to a density of 152 micromol Cu2+ ions/g gel was characterised by frontal analysis experiments using horse heart myoglobin (HMYO) at pH 7.0 and pH 9.0. From the derived isotherms, the adsorption capacity, q(m), for the binding of HMYO to immobilised Cu2+-N-(2-pyridylmethyl)aminoacetate (im-Cu2+-carbpyr)-Sepharose CL-4B at these pH values was found to be 1.92 and 1.91 micromol/g sorbent, respectively, whilst the dissociation constants K(D) were 0.0092 x 10(-6) M and 0.0062 x 10(-6) M at pH 7.0 and pH 9.0, respectively, indicating that the HMYO-im-Cu2+-N-(2-pyridylmethyl)aminoacetate complex was more stable under alkaline conditions, although the binding capacity in terms of micromol protein/g gel remained essentially unchanged. The selectivity features of the im-Cu2+-carbpyr-Sepharose CL-4B sorbent were further characterised in terms of the binding properties with several human serum proteins at pH 5.0, pH 7.0 and pH 9.0.

Adsorption↗
PubMed1999

Virtual microscopy: high resolution digital photomicrography as a tool for light microscopy simulation.

Recent advances in microcomputers and high resolution digital video cameras provide pathologists the opportunity to combine precision optics with digital imaging technology and develop new educational and research tools. We review recent advances in virtual microscopy and describe techniques for viewing digital images using a microcomputer-based workstation to simulate light microscopic examination, including scanning at low power to select features of interest and zooming to increase magnification. Hardware and software components necessary to acquire digital images of histological and cytological slides, and closely simulate their examination under a light microscope are discussed. The workstation is composed of a MicroLumina digital scanning camera (Leaf Systems, Southborough, MA), light microscope (Olympus Optical Co., Lake Success, NY), Pentium (Intel Corp., Santa Clara, CA) 166 MHz microcomputer configured with 64 megabytes of random access memory (RAM), a MGA Millenium Powerdesk graphics card (Matrox Graphics, Inc., Montreal, Canada) and Photoshop software (Adobe Systems Inc., San Jose, CA) running in a Windows 95 (Microsoft Corp., Redmond, WA) environment. Images with spatial resolutions of up to 2700 x 3400 pixels in 36-bit color, can be displayed simultaneously as distinct images in a montage, or merged into a single composite image file to highlight significant features of a histological or cytological slide. These image files are saved in Joint Photographers Experts Group (JPEG) format using compression ratios of up to 80:1 without detectable visual degradation. The advantages and technical limitations of various workstation components are addressed and applications of this technology for pathology education, proficiency testing, telepathology, and database development are discussed.

Forecasting↗
PubMed2003

Efficient and accurate experimental design for enzyme kinetics: Bayesian studies reveal a systematic approach.

In areas such as drug development, clinical diagnosis and biotechnology research, acquiring details about the kinetic parameters of enzymes is crucial. The correct design of an experiment is critical to collecting data suitable for analysis, modelling and deriving the correct information. As classical design methods are not targeted to the more complex kinetics being frequently studied, attention is needed to estimate parameters of such models with low variance. We demonstrate that a Bayesian approach (the use of prior knowledge) can produce major gains quantifiable in terms of information, productivity and accuracy of each experiment. Developing the use of Bayesian Utility functions, we have used a systematic method to identify the optimum experimental designs for a number of kinetic model data sets. This has enabled the identification of trends between kinetic model types, sets of design rules and the key conclusion that such designs should be based on some prior knowledge of K(M) and/or the kinetic model. We suggest an optimal and iterative method for selecting features of the design such as the substrate range, number of measurements and choice of intermediate points. The final design collects data suitable for accurate modelling and analysis and minimises the error in the parameters estimated.

Bayes Theorem↗
PubMed2002

Histopathologic differences due to bacterial species in acute otitis media.

OBJECTIVE: To compare selected features of histopathology in acute otitis media caused by various bacteria and examine potential differences due to bacterial species, as well as possible correlation to experimental and human clinical findings. METHODS: Rat models of acute otitis media caused by Streptococcus pneumoniae (MC), non-typeable or type b Haemophilus influenzae (NTHI/HIB) or Moraxella catarrhalis (MC) were studied longitudinally up to 6 months after bacterial challenge. Findings related to dynamics of goblet cell density, modeling and remodeling of bone tissue structures and polyp, as well as fibrous adhesion formation and persistence are presented. RESULTS: Middle ear goblet cell density progressed to peak 2 weeks after bacterial inoculation, thereafter gradually normalizing. However, density and accordingly middle ear secretory capacity was still significantly increased after 6 months in all bacteria, except MC. The HI species induced the highest increase. Initial osteoresorption was followed by massive osteoneogenesis, progressing to a peak after 2-3 months, followed by some degree of normalization, concurrently classic remodeling. Primarily SP, but also the HI species induced more new bone formation than MC. Mucosal polyp and fibrous adhesion formation occurred regardless of bacterial species. Most polyps appeared in the early phases and the HI species induced formation of more polyps and adhesions than the other bacteria. CONCLUSION: Acute middle ear infection with the Haemophilus species induce the highest increase of mucosal secretory capacity, lasting for at least 6 months after the acute incident. Thus, a subsequent development of secretory otitis media seems more likely following infection with these bacteria. Equivalently, mucosal scarring observed as polyp and fibrous adhesion formation was more severe following Haemophilus infection. S. pneumoniae induced the most marked changes of bone tissue structures, seen as initial osteoresorption and subsequent osteoneogenesis. Overall, infection with M. catarrhalis induced the mildest changes.

Acute Disease↗
PubMed2001

Image matching algorithms for breech face marks and firing pins in a database of spent cartridge cases of firearms.

On the market several systems exist for collecting spent ammunition data for forensic investigation. These databases store images of cartridge cases and the marks on them. Image matching is used to create hit lists that show which marks on a cartridge case are most similar to another cartridge case. The research in this paper is focused on the different methods of feature selection and pattern recognition that can be used for optimizing the results of image matching. The images are acquired by side light images for the breech face marks and by ring light for the firing pin impression. For these images a standard way of digitizing the images used. For the side light images and ring light images this means that the user has to position the cartridge case in the same position according to a protocol. The positioning is important for the sidelight, since the image that is obtained of a striation mark depends heavily on the angle of incidence of the light. In practice, it appears that the user positions the cartridge case with +/-10 degrees accuracy. We tested our algorithms using 49 cartridge cases of 19 different firearms, where the examiner determined that they were shot with the same firearm. For testing, these images were mixed with a database consisting of approximately 4900 images that were available from the Drugfire database of different calibers.In cases where the registration and the light conditions among those matching pairs was good, a simple computation of the standard deviation of the subtracted gray levels, delivered the best-matched images. For images that were rotated and shifted, we have implemented a "brute force" way of registration. The images are translated and rotated until the minimum of the standard deviation of the difference is found. This method did not result in all relevant matches in the top position. This is caused by the effect that shadows and highlights are compared in intensity. Since the angle of incidence of the light will give a different intensity profile, this method is not optimal. For this reason a preprocessing of the images was required. It appeared that the third scale of the "à trous" wavelet transform gives the best results in combination with brute force. Matching the contents of the images is less sensitive to the variation of the lighting. The problem with the brute force method is however that the time for calculation for 49 cartridge cases to compare between them, takes over 1 month of computing time on a Pentium II-computer with 333MHz. For this reason a faster approach is implemented: correlation in log polar coordinates. This gave similar results as the brute force calculation, however it was computed in 24h for a complete database with 4900 images.A fast pre-selection method based on signatures is carried out that is based on the Kanade Lucas Tomasi (KLT) equation. The positions of the points computed with this method are compared. In this way, 11 of the 49 images were in the top position in combination with the third scale of the à trous equation. It depends however on the light conditions and the prominence of the marks if correct matches are found in the top ranked position. All images were retrieved in the top 5% of the database. This method takes only a few minutes for the complete database if, and can be optimized for comparison in seconds if the location of points are stored in files. For further improvement, it is useful to have the refinement in which the user selects the areas that are relevant on the cartridge case for their marks. This is necessary if this cartridge case is damaged and other marks that are not from the firearm appear on it.

Algorithms↗
PubMed2001

Comparing Bayesian neural network algorithms for classifying segmented outdoor images.

In this paper we investigate the Bayesian training of neural networks for region labelling of segmented outdoor scenes; the data are drawn from the Sowerby Image Database of British Aerospace. Neural networks are trained with two Bayesian methods, (i) the evidence framework of MacKay (1992a,b) and (ii) a Markov Chain Monte Carlo method due to Neal (1996). The performance of the two methods is compared to evaluating the empirical learning curves of neural networks trained with the two methods. We also investigate the use of the Automatic Relevance Determination method for input feature selection.

Algorithms↗