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 1,459 records · Page 81Linked to original sources
PubMed1998

Temporal and spatial features of slow positive potential related to visual selective response during head-down-tilt.

The temporal and spatial features of visual ERP changes during 2 hour HDT (-10 degrees) were compared with that during HUT (+20 degrees) in 15 normal subjects. The stimuli consisted of two color LED flashes appeared randomly in left (LVF) or right (RVF) visual field with the same probability. The subjects were asked to make switch response to target signals (T) differentially: switching to left for T in LVF and to right for T in RVF, and making no response to non-target signals (N). Five sets of tests were made during HUT and HDT. ERPs were obtained from 9 locations on the scalp. Mean amplitude of the ERPs in the period 0.30-0.55 s were taken as the amplitude of slow positive potential (P400). The results obtained were: (1) The mean amplitude of P400 decreased during HDT which was more significant at the 2nd, 3rd and 5th set of tests; (2) spatially, the reduction of mean P400 amplitude during HDT was more significant for signals from RVF and were more significant at posterior and central brain regions than those from frontal locations. These data provide further evidence showing that the higher brain function was affected by the simulated weightlessness and that this effect was not only transient but also with interesting spatial characteristics.

Adult↗
PubMedDate not supplied

Second-order features extraction in the cat visual cortex: selective and invariant sensitivity of neurons to the shape and orientation of crosses and corners.

About 1/3 of neurons (52/174) studied in the cat striate cortex (area 17 or V1) gave a larger (by 3.2 times on average) response to a flashed cross or corner centered in the receptive field (RF) than to a single bar of optimal orientation. Most such neurons (71.4%) were found to be highly selective both to shape (angle between the lines) and to orientation of these figures. In the studied neuronal selection we also found all possible types of invariance of sensitivity to orientation and/or shape of these figures. We found neurons with selectivity to form of the figure and invariance to its orientation and vice versa. Some cells were found invariant both to form and orientation of the cross or corner but highly sensitive to flashing of any such figure in the RF. The role of RF center and surrounding area in sensitivity to cross figures was also studied in 44 additional units. Separated and combined stimulation of these zones revealed in different units summation, antagonism and absence of interaction of these zones by the selectivity index (figure/bar response ratio). Possible mechanisms of the described effects are discussed as well as their functional implication for second-order feature extraction in the visual cortex: selective or invariant sensitivity of neurons to the shape and orientation of the line-crossings.

Animals↗
PubMed2003

Human-mouse gene identification by comparative evidence integration and evolutionary analysis.

The identification of genes in the human genome remains a challenge, as the actual predictions appear to disagree tremendously and vary dramatically on the basis of the specific gene-finding methodology used. Because the pattern of conservation in coding regions is expected to be different from intronic or intergenic regions, a comparative computational analysis can lead, in principle, to an improved computational identification of genes in the human genome by using a reference, such as mouse genome. However, this comparative methodology critically depends on three important factors: (1) the selection of the most appropriate reference genome. In particular, it is not clear whether the mouse is at the correct evolutionary distance from the human to provide sufficiently distinctive conservation levels in different genomic regions, (2) the selection of comparative features that provide the most benefit to gene recognition, and (3) the selection of evidence integration architecture that effectively interprets the comparative features. We address the first question by a novel evolutionary analysis that allows us to explicitly correlate the performance of the gene recognition system with the evolutionary distance (time) between the two genomes. Our simulation results indicate that there is a wide range of reference genomes at different evolutionary time points that appear to deliver reasonable comparative prediction of human genes. In particular, the evolutionary time between human and mouse generally falls in the region of good performance; however, better accuracy might be achieved with a reference genome further than mouse. To address the second question, we propose several natural comparative measures of conservation for identifying exons and exon boundaries. Finally, we experiment with Bayesian networks for the integration of comparative and compositional evidence.

Animals↗
PubMed2004

Mu rhythm-based cursor control: an offline analysis.

OBJECTIVE: To classify the EEG data recorded in mu rhythm-based cursor control experiments with 4 possible choices. METHODS: The algorithm included preprocessing, feature extraction, and classification. Two spatial filters, common average reference and common spatial subspace decomposition, were used in preprocessing to improve the signal-to-noise ratio, and then two features were extracted based on the power spectrum and the time course of the mu rhythm respectively. A Fisher ratio was defined to select channels in feature extraction. A 2-dimensional linear classifier was trained for final classification. RESULTS: Two types of classifiers were trained for the training dataset. The uniform classifier gave a classification accuracy of 76.4%, and the classifier trained by leave-one-out method gave a classification accuracy of 74.4%, both higher than the online accuracy 69.5%. The uniform classifier was applied to the test dataset and the classification accuracy was 65.9%, lower than the online accuracy 73.2%. CONCLUSIONS: Spatial filtering can give a notable improvement in classification accuracy. The time course of the mu rhythm, as well as the power of the mu rhythm, shows difference between the 4 targets, and can contribute to the classification. SIGNIFICANCE: The spatial filtering, feature extraction and channel selection methods in the algorithm will provide some practical suggestions for further study on the mu rhythm-based brain-computer interface.

Algorithms↗
PubMed2001

Vocal syntax development in the white-crowned sparrow (Zonotrichia leucophrys).

Birdsong development exemplifies the interplay between experience and predisposition that occurs during behavioral ontogeny. Songbirds must hear song models to develop normal song, yet they preferentially learn conspecific song when given a choice in the laboratory. To the extent that features guiding this selective learning are pre-encoded in the brain, such features should also develop in the song of young birds not exposed to them in tutor models. To investigate whether song syntax-phrase number and order--is such a feature in the white-crowned sparrow (Zonotrichia leucophrys), the authors tutored males of this species with separate phrase models. Birds learned and assembled these into songs of species-typical sequence, suggesting that syntax is to some degree pre-encoded in white-crowned sparrows. Birds also learned heterospecific phrases, confirming previous evidence that note phonology is not the primary cue for selective song learning in this species.

Animals↗
PubMed2006

Evidence for an elders' advantage in the naive product usability judgments of older and younger adults.

OBJECTIVE: To determine whose naive judgments of consumer product usability are more accurate--those of younger or older adults. Accuracy is here defined as judgments compatible with results from performance-based usability tests. BACKGROUND: Older adults may be better able to predict usability problems than younger adults, making them particularly good participants in studies contributing to the user-centered design of products. This advantage, if present, may stem from older adults' motivation for more usable products or from their experience adapting their own environments to meet their changing physical, cognitive, and sensory needs. METHOD: Sixty older participants (ages 65-75 years) and 60 younger ones (ages 18-22 years) evaluated illustrations of consumer products on specific criteria (e.g., readability, learnability, or error rates). They either rated a single design for each product or ranked six alternative designs. They also explained their choices, indicated which features were most critical for usability, and selected usability-enhancing modifications. RESULTS: Although there was no reliable age difference in the amount of usability information provided in the open-ended explanations, older adults were more accurate at ranking alternative designs, selecting the most usability-critical features, and selecting usability-enhancing modifications (all ps < .05). CONCLUSION: The usability judgments of older adults are more accurate than those of younger adults when these judgments are solicited in a fixed-alternative, but not open-ended, format. APPLICATION: Because older adults are more discerning about potential product usability problems, they may be particularly valuable as research participants in early-stage design research (prior to the availability of working prototypes).

Adolescent↗
PubMed1986

Operation and performance of an automatic metaphase finder based on the MRC fast interval processor.

The Medical Research Council's fast interval processor (FIP) has been adapted for metaphase finding and selection. This article summarizes recent improvements to the hardware, and describes the selection of image features. The system uses a highly simplified but effective clustering procedure to reduce computation time, and incorporates a ranking algorithm based on computed cluster features so that high-quality metaphases can be preferentially selected. Experimental results indicate that the system can detect high-quality metaphases rapidly in "rich" material and a high proportion of the available metaphases in "sparse" material. It can handle a wide range of material with good repeatability of performance.

Automation↗
PubMed1997

An epitope tagged mammalian/prokaryotic expression vector with positive selection of cloned inserts.

A dual eukaryotic/prokaryotic expression vector has been developed which combines the features of positive selection for cloned inserts along with the production of an epitope-tagged cDNA insert by transient transfection in mammalian cells as well as high level induced expression in E. coli cells harbouring T7 RNA polymerase. This vector, pZilch, has two MCSs flanking a mutant E. coli phenylalanyl-tRNA synthetase gene, pheS, which when expressed in combination with the phenylalanine analog p-CI-Phe, results in termination of host cell protein synthesis. Cloning of inserts using unique sites in the flanking MCS regions results in loss of the pZilch pheS allele and hence permits growth of colonies harbouring recombinants on p-Cl-Phe plates. Additional features of the vector include an optimal Kozak consensus sequence for high level eukaryotic cell expression and an efficient prokaryotic translation initiation site in frame and downstream from the eukaryotic initiation site. Recombinant proteins can be produced with an N-terminal FLAG epitope which can be removed via a specific protease cleavage site. Flanking T7 and SP6 RNA polymerase promoter sites permit in vitro transcription and translation of cloned inserts. A derivative of the vector has also been constructed enabling nuclear accumulation of the tagged proteins via an SV40 nuclear localisation signal upstream of the 5' MCS.

Amino Acid Sequence↗
PubMed2004

Subsets with restricted immunoglobulin gene rearrangement features indicate a role for antigen selection in the development of chronic lymphocytic leukemia.

We recently identified a chronic lymphocytic leukemia (CLL) subgroup using the immunoglobulin variable heavy-chain (V(H)) gene V(H)3-21 with almost identical heavy-chain complementarity determining region 3s (HCDR3s) and preferential variable light-chain (V(L)) gene usage, suggesting recognition of a common antigen epitope in this subset. To further explore the B-cell receptors (BCRs) in CLL, we characterized 407 V(H) rearrangements amplified from 346 CLLs regarding V(H), diversity (D), and joining (J(H)) gene usage and performed multiple alignment of the HCDR3 sequences. These analyses revealed 3 small subsets (2 V(H)1-69 groups, 7 cases; and 1 V(H)1-2 group, 5 cases) with highly restricted HCDR3 features including identical V(H)/D/J(H) usage, HCDR3 lengths, and shared N-sequences, in addition to the V(H)3-21 group (22 cases). Furthermore, another 3 groups (9 V(H)1-3(+) cases, 3 V(H)1-18(+) cases, and 5 V(H)4-39(+) cases) had essentially identical V(H)/D/J(H) use and similar HCDR3 lengths but less conserved N-regions. Analysis in all 6 of these subgroups showed restriction in V(L) gene use, whereas no association between V(H) and V(L) usage was found in cases without HCDR3 similarities. Altogether, structurally similar HCDR3s associated with preferential V(L) gene usage implies selection of BCRs, especially in subsets showing high HCDR3 similarities, thus pointing to restricted antigen recognition sites and possibly involvement of specific antigens in CLL development.

Adult↗
PubMed2003

Pronuclear and morphological features as a cumulative score to select embryos in ICSI (intracytoplasmic sperm injection) cycles according to sperm origin.

PURPOSE: To analyze best parameter to select embryos according to sperm origin in ICSI cycles. METHODS: One hundred seventy-two ICSI cycles were divided among three different groups: A (ejaculated spermatozoa from nonmale factor infertility), B (ejaculated spermatozoa from oligospermia), and C (spermatozoa from azoospermia). Embryos were divided on Day 1 into two patterns: S0 (pronuclei (PN) aligned and close with normal arrangement of nucleoli) and S1 (when these characteristics were absent) and also on transfer day according to morphological features. RESULTS: Relationships of PN patterns related to sperm origin were noted. More S0 embryos were detected with better sperm quality. Higher number of good quality embryos was obtained when male factor was absent. Ejaculated and epididymal spermatozoa provide better quality embryos than do testicular spermatozoa. CONCLUSIONS: PN classification associated with transfer day morphology is valuable additional noninvasive criterion for elective embryo transfer, mainly in the cases with severe male factor.

Abortion, Spontaneous↗
PubMed2003

Evaluation of the forearm EMG signal features for the control of a prosthetic hand.

The purpose of this research is to select the best features to have a high rate of motion classification for controlling an artificial hand. Here, 19 EMG signal features have been taken into account. Some of the features suggested in this study include combining wavelet transform with other signal processing techniques. An assessment is performed with respect to three points of view: (i) classification of motions, (ii) noise tolerance and (iii) calculation complexity. The energy of wavelet coefficients of EMG signals in nine scales, and the cepstrum coefficients were found to produce the best features in these views.

Artificial Limbs↗
PubMed1985

Spontaneous resolution of cervical warty atypia: the relevance of clinical and nuclear DNA features: a prospective study.

A selected group of 51 patients with histological features of cervical warty atypia was observed for 1 year. Initial features were compared with a cohort of 50 patients with CIN. Twenty-four patients showed resolution of the warty atypia, in 19 it remained unchanged, and in eight patients CIN or microinvasive disease was present on repeat biopsy. The DNA patterns obtained by tissue imprint of the initial biopsies and subsequent integrated densitometry, using a Vickers M85 scanning densitometer, showed that warty atypias could be objectively distinguished from CIN, but that the method could not predict the group of warty atypias that would resolve. There was no clinical feature or group of features which allowed a means of predicting the outcome of such lesions.

Adult↗
PubMed2005

Improving functional annotation of non-synonomous SNPs with information theory.

Automated functional annotation of nsSNPs requires that amino-acid residue changes are represented by a set of descriptive features, such as evolutionary conservation, side-chain volume change, effect on ligand-binding, and residue structural rigidity. Identifying the most informative combinations of features is critical to the success of a computational prediction method. We rank 32 features according to their mutual information with functional effects of amino-acid substitutions, as measured by in vivo assays. In addition, we use a greedy algorithm to identify a subset of highly informative features. The method is simple to implement and provides a quantitative measure for selecting the best predictive features given a set of features that a human expert believes to be informative. We demonstrate the usefulness of the selected highly informative features by cross-validated tests of a computational classifier, a support vector machine (SVM). The SVM's classification accuracy is highly correlated with the ranking of the input features by their mutual information. Two features describing the solvent accessibility of "wild-type" and "mutant" amino-acid residues and one evolutionary feature based on superfamily-level multiple alignments produce comparable overall accuracy and 6% fewer false positives than a 32-feature set that considers physiochemical properties of amino acids, protein electrostatics, amino-acid residue flexibility, and binding interactions.

Analysis of Variance↗
PubMed2006

Fibre selection based on an overall analytical feature comparison for the solid-phase microextraction of trihalomethanes from drinking water.

This paper describes the optimization of solid-phase microextraction (SPME) conditions for three different fibres (Carboxen-polydimethylsiloxane (CAR-PDMS), divinylbenzene-Carboxen-polydimethylsiloxane (DVB-CAR-PDMS) and polydimethylsiloxane-divinylbenzene (PDMS-DVB)) used to determine trihalomethanes (THMs) in water by headspace solid-phase microextraction and gas chromatography (HS-SPME-GC). The influence of temperature and salting-out effect was examined using a central composite design for each fibre. Extraction time was studied separately at the optimum values found for temperature and sodium chloride concentration (40 degrees C and 0.36g mL-1). The HS-SPME-GC-MS method for each fibre was characterised in terms of linearity, detection (LOD) and quantification (LOQ) limits and repeatability. The fibre PDMS-DVB was selected as it provided a broader linear range, better repeatability and lower detection and quantification limits than the others, particularly CAR-PDMS fibre. The accuracy of the proposed method using the PDMS-DVB fibre was checked by a recovery study in both ultrapure and tap water. A blank analysis study showed the absence of memory effects for this fibre. The reproducibility (expressed as a percentage of relative standard deviation) was 6-11% and the detection limits were between 0.078 and 0.52microgL-1 for bromoform and chloroform, respectively. Finally, the method was applied to determine THM concentration in two drinking water samples.

Gas Chromatography-Mass Spectrometry↗
PubMed1997

Mapping features of HIV-1 integrase near selected sites on viral and target DNA molecules in an active enzyme-DNA complex by photo-cross-linking.

The virally encoded integrase protein carries out retroviral integration, and to do so, it must make specific interactions with both viral and target DNA sequences. The retroviral integrase has three domains: an amino-terminal region of about 50 amino acids that contains a zinc finger-like motif, a tightly folded, phylogenetically conserved core domain that contains the active site, and a carboxy-terminal domain that can bind DNA in a nonspecific manner. The complete roles of the amino- and carboxyl-terminal domains have not yet been determined, but they appear to participate in multimerization and nonspecific or target DNA binding, respectively. The number and identity of integrase's DNA binding sites have been difficult to determine by conventional mutagenesis studies. In this report, we describe a photo-cross-linking approach to address these issues. Our findings suggest that HIV-1 integrase contacts with conserved features of the viral DNA end are likely to be mediated by residues in two peptides within the conserved core domain. Additional cross-links were seen between viral DNA and the carboxy-terminal DNA binding domain. Numerous sites in integrase, including peptides in each of the three domains, could be cross-linked to target DNA features. Integrase is known to function as a multimer, and it remains to be determined which specific contacts are in cis or trans with respect to the active site.

Azides↗
PubMed2006

Whole plant responses, key processes, and adaptation to drought stress: the case of rice.

Most high-yielding rice cultivars developed for irrigated conditions, including the widely grown lowland variety IR64, are highly susceptible to drought stress. This limits their adoption in rainfed rice environments where there is a risk of water shortage during the growing season. Mapping studies using lowland-by-upland rice populations have provided limited information about the genetic basis of variation in yield under drought. One approach to simultaneously improve and understand rice drought tolerance is to generate backcross populations, select superior lines in managed stress environments, and then evaluate which features of the selected lines differ from the recurrent parent. This approach was been taken with IR64, using a range of tolerant and susceptible cultivars as donor parents. Yields of the selected lines measured across 13 widely contracting water environments were generally greater than IR64, but genotype-by-environment effects were large. Traits expected to vary between IR64 and selected lines are plant height, because many donors were not semi-dwarf types, and maturity, because selection in a terminal stress environment is expected to favour earliness. In these experiments it was found that some lines that performed better under upland drought were indeed taller than IR64, but that shorter lines with good yield under drought could also be identified. In trials where drought stress developed in previously flooded (lowland) fields, height was not associated with performance. There was little change in maturity with selection. Other notable differences between IR64 and the selected backcross lines were in their responses to applied ABA and ethylene in greenhouse experiments at the vegetative stage and in leaf rolling observed under chronic upland stress in the field. These observations are consistent with the hypothesis that adaptive responses to drought can effectively allow for improved performance across a broad range of water environments. The results indicate that the yield of IR64 under drought can be significantly improved by backcrossing with selection under stress. In target environments where drought is infrequent but significant in certain years, improved IR64 with greater drought tolerance would be a valuable option for farmers.

Adaptation, Physiological↗
PubMed2004

Neuronal synchronization and selective color processing in the human brain.

In the present study, subjects selectively attended to the color of checkerboards in a feature-based attention paradigm. Induced gamma band responses (GBRs), the induced alpha band, and the event-related potential (ERP) were analyzed to uncover neuronal dynamics during selective feature processing. Replicating previous ERP findings, the selection negativity (SN) with a latency of about 160 msec was extracted. Furthermore, and similarly to previous EEG studies, a gamma band peak in a time window between 290 and 380 msec was found. This peak had its major energy in the 55- to 70-Hz range and was significantly larger for the attended color. Contrary to previous human induced gamma band studies, a much earlier 40- to 50-Hz peak in a time window between 160 and 220 msec after stimulus onset and, thus, concurrently to the SN was prominent with significantly more energy for attended as opposed to unattended color. The induced alpha band (9.8-11.7 Hz), on the other hand, exhibited a marked suppression for attended color in a time window between 450 and 600 msec after stimulus onset. A comparison of the time course of the 40- to 50-Hz and 55- to 70-Hz induced GBR, the induced alpha band, and the ERP revealed temporal coincidences for changes in the morphology of these brain responses. Despite these similarities in the time domain, the cortical source configuration was found to discriminate between induced GBRs and the SN. Our results suggest that large-scale synchronous high-frequency brain activity as measured in the human GBR play a specific role in attentive processing of stimulus features.

Adult↗
PubMed1990

A principle for the formation of the spatial structure of cortical feature maps.

Orientation-selective cells in the striate cortex of higher animals are organized as a hierarchical topographic map of two stimulus features: (i) position in visual space and (ii) orientation. We show that the observed structure of the topographic map can arise from a principle of continuous mapping. For the realization of this principle we use a mathematical model that can be interpreted as an adaptive process changing a set of synaptic weights, or synaptic connection strengths, between two layers of cells. The patterns of orientation preference and selectivity generated by the model are similar to the patterns seen in the visual cortex of macaque monkey and cat and correspond to a neural projection that maps a more than two-dimensional feature space onto a two-dimensional cortical surface under the constraint that shape and position of the receptive fields of the neurons very smoothly over the cortical surface.

Animals↗