Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “unsupervised clustering”

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 559 records · Page 31Linked to original sources
PubMed1997

Self-organizing neural maps of the coding sequences of G-protein-coupled receptors reveal local domains associated with potentially functional determinants in the proteins.

Mapping of the coding sequences of the best characterized subfamilies of G-protein-coupled receptors is performed with unsupervised neural networks based on a winner-take-all strategy. High order features therefrom extracted originate signals along the aligned protein sequences of the different subfamilies. These plots reveal characteristic domains common and/or characteristic of the receptor subfamily. By comparison with the existing experimental results, it is obtained that most of the regions signalled by clustering overlap with possible functional regions in the folded proteins. This is particularly noticeable for the third cytoplasmic loop, which is likely to be involved in the molecular coupling with the G-proteins. The results suggest that functional regions in proteins may be characterized by intrinsic representative features in the coding sequences which can be enlighted by high order mapping.

Binding Sites↗
PubMed2006

The relationship between unsupervised time after school and physical activity in adolescent girls.

BACKGROUND: Rising obesity and declining physical activity levels are of great concern because of the associated health risks. Many children are left unsupervised after the school day ends, but little is known about the association between unsupervised time and physical activity levels. This paper seeks to determine whether adolescent girls who are without adult supervision after school are more or less active than their peers who have a caregiver at home. METHODS: A random sample of girls from 36 middle schools at 6 field sites across the U.S. was selected during the fall of the 2002-2003 school year to participate in the baseline measurement activities of the Trial of Activity for Adolescent Girls (TAAG). Information was collected using six-day objectively measured physical activity, self-reported physical activity using a three-day recall, and socioeconomic and psychosocial measures. Complete information was available for 1422 out of a total of 1596 respondents.Categorical variables were analyzed using chi square and continuous variables were analyzed by t-tests. The four categories of time alone were compared using a mixed linear model controlling for clustering effects by study center. RESULTS: Girls who spent more time after school (> or = 2 hours per day, > or = 2 days per week) without adult supervision were more active than those with adult supervision (p = 0.01). Girls alone for > or = 2 hours after school, > or = 2 days a week, on average accrue 7.55 minutes more moderate to vigorous physical activity (MVPA) per day than do girls who are supervised (95% confidence interval ([C.I]). These results adjusted for ethnicity, parent's education, participation in the free/reduced lunch program, neighborhood resources, or available transportation. Unsupervised girls (n = 279) did less homework (53.1% vs. 63.3%), spent less time riding in a car or bus (48.0% vs. 56.6%), talked on the phone more (35.5% vs. 21.1%), and watched more television (59.9% vs. 52.6%) than supervised girls (n = 569). However, unsupervised girls also were more likely to be dancing (14.0% vs. 9.3%) and listening to music (20.8% vs. 12.0%) (p < .05). CONCLUSION: Girls in an unsupervised environment engaged in fewer structured activities and did not immediately do their homework, but they were more likely to be physically active than supervised girls. These results may have implications for parents, school, and community agencies as to how to structure activities in order to encourage teenage girls to be more physically active.

Journal Article↗
PubMed2005

Systematic association of genes to phenotypes by genome and literature mining.

One of the major challenges of functional genomics is to unravel the connection between genotype and phenotype. So far no global analysis has attempted to explore those connections in the light of the large phenotypic variability seen in nature. Here, we use an unsupervised, systematic approach for associating genes and phenotypic characteristics that combines literature mining with comparative genome analysis. We first mine the MEDLINE literature database for terms that reflect phenotypic similarities of species. Subsequently we predict the likely genomic determinants: genes specifically present in the respective genomes. In a global analysis involving 92 prokaryotic genomes we retrieve 323 clusters containing a total of 2,700 significant gene-phenotype associations. Some clusters contain mostly known relationships, such as genes involved in motility or plant degradation, often with additional hypothetical proteins associated with those phenotypes. Other clusters comprise unexpected associations; for example, a group of terms related to food and spoilage is linked to genes predicted to be involved in bacterial food poisoning. Among the clusters, we observe an enrichment of pathogenicity-related associations, suggesting that the approach reveals many novel genes likely to play a role in infectious diseases.

Bacteria↗
PubMed2006

Molecular characterization of human adenomyosis.

Adenomyosis is a common gynaecological disorder characterized by the abnormal growth of endometrium into the myometrium and myometrial hypertrophy/hyperplasia. Uterine fibroids are benign neoplasms of the myometrium, and they represent a diagnostic pitfall for adenomyosis. In this study, we have used the genome-wide Affymetrix U133 Plus 2.0 microarray platform to compare the gene expression patterns of adenomyosis, uterine fibroids, normal endometrium and myometrium. Unsupervised principal component analysis (PCA) revealed that these four tissue types could be segregated from one another solely based on their gene expression profiles. Analysis of variance (ANOVA), followed by Tukey means separation test, significance analysis of microarrays (SAM) and 2-fold change threshold, identified 7415 probe sets as differentially expressed among the four groups of samples. Supervised cluster analysis based on these probe sets clustered adenomyosis most closely with endometrium and uterine fibroids with myometrium, consistent with the anatomic origin of these two diseases. The Tukey means separation post hoc testing found 2073 probe sets altered between adenomyosis and normal endometrium or myometrium, and 2327 probe sets altered in expression when comparing uterine fibroids with myometrium. Using Ingenuity Pathways Analysis (IPA), we found 9 highly significant functional networks in adenomyosis and 10 in uterine fibroids. Notably, the top network in both cases was associated with functions implicated in cancer and cell death. Finally, we compared the gene expression profiles of adenomyosis and uterine fibroids and identified 471 differentially expressed probe sets that may represent potential biomarkers for the differential diagnosis of these diseases.

Analysis of Variance↗
PubMed2004

Molecular prediction of response to 5-fluorouracil and interferon-alpha combination chemotherapy in advanced hepatocellular carcinoma.

PURPOSE: The prognosis of hepatocellular carcinoma (HCC) is very poor, particularly in patients with tumors that have invaded the major branches of the portal vein. Combination chemotherapy with intra-arterial 5-fluorouracil and subcutaneous interferon-alpha has shown promising results for such advanced HCC, but it is important to develop the ability to accurately predict chemotherapeutic responses. EXPERIMENTAL DESIGN: We analyzed the expression of 3,080 genes using a polymerase chain reaction-based array in 20 HCC patients who were treated with combination chemotherapy after reduction surgery. After unsupervised analyses, a supervised classification method for predicting chemotherapeutic responses was constructed. To minimize the number of predictive genes, we used a random permutation test to select only significant (P < 0.01) genes. A leave-one-out cross-validation confirmed the gene selection. We also prepared an additional 11 cases for validation of predictive performance. RESULTS: Hierarchical clustering analysis and principal component analysis with all 3,080 genes revealed distinct gene expression patterns in responders (those with complete response or partial response) and nonresponders (those with stable disease or progressive disease) to the combination chemotherapy. Using a weighted-voting classification method with either all genes or only significant genes as assessed by permutation testing, the objective responses to treatment were correctly predicted in 17 of 20 cases (accuracy, 85%; positive predictive value, 100%; negative predictive value, 80%). Moreover, patients in the validation dataset could be classified into two distinct prognostic groups using 63 predictive genes. CONCLUSIONS: Molecular analysis of 63 genes can predict the response of patients with advanced HCC and major portal vein tumor thrombi to combination chemotherapy with 5-fluorouracil and interferon-alpha.

Adult↗
PubMed2004

Gene expression profiles in primary ovarian serous papillary tumors and normal ovarian epithelium: identification of candidate molecular markers for ovarian cancer diagnosis and therapy.

With the goal of identifying genes with a differential pattern of expression between ovarian serous papillary carcinomas (OSPCs) and normal ovarian (NOVA) epithelium and using this knowledge for the development of novel diagnostic and therapeutic markers for ovarian cancer, we used oligonucleotide microarrays with probe sets complementary to 12,533 genes to analyze the gene expression profiles of 10 primary OSPC cell lines, 2 established OSPC cell lines (UCI-101, UCI-107) and 5 primary NOVA epithelial cultures. Unsupervised analysis of gene expression data identified 129 and 170 genes that exhibited >5-fold upregulation and downregulation, respectively, in primary OSPC compared to NOVA. Genes overexpressed in established OSPC cell lines had little correlation with those overexpressed in primary OSPC, highlighting the divergence of gene expression that occurs as a result of long-term in vitro growth. Hierarchical clustering of the expression data readily distinguished normal tissue from primary OSPC. Laminin, claudin 3, claudin 4, tumor-associated calcium signal transducers 1 and 2 (TROP-1/Ep-CAM, TROP-2), ladinin 1, S100A2, SERPIN2 (PAI-2), CD24, lipocalin 2, osteopontin, kallikrein 6 (protease M), kallikrein 10, matriptase (TADG-15) and stratifin were among the most highly overexpressed genes in OSPC compared to NOVA. Downregulated genes in OSPC included transforming growth factor-beta receptor III, platelet-derived growth factor receptor alpha, SEMACAP3, ras homolog gene family member I (ARHI), thrombospondin 2 and disabled-2/differentially expressed in ovarian carcinoma 2 (Dab2/DOC2). Differential expression of some of these genes, including claudin 3, claudin 4, TROP-1 and CD24, was validated by quantitative RT-PCR and flow cytometry on primary OSPC and NOVA. Immunohistochemical staining of formalin-fixed, paraffin-embedded tumor specimens from which primary OSPC cultures were derived further confirmed differential expression of CD24 and TROP-1/Ep-CAM markers on OSPC vs. NOVA. These results, obtained with highly purified primary cultures of ovarian cancer, highlight important molecular features of OSPC and may provide a foundation for the development of new type-specific therapies against this disease.

Adult↗
PubMed2004

Ischemic but not pharmacological preconditioning elicits a gene expression profile similar to unprotected myocardium.

Pharmacological (PPC) and ischemic preconditioning (IschPC) provide comparable protection against ischemia in the heart. However, the genomic phenotype may depend on the type of preconditioning. Isolated perfused rat hearts were used to evaluate transcriptional responses to PPC and IschPC in the presence (mediator/effector response) or absence (trigger response) of 40 min of test ischemia using oligonucleotide microarrays. IschPC was induced by 3 cycles of 5 min of ischemia, and PPC by 15 min of 2.1 vol% isoflurane. Unsupervised analysis methods were used to identify gene expression patterns. PPC and IschPC were accompanied by marked alterations in gene expression. PPC and IschPC shared only approximately 25% of significantly up- and downregulated genes after triggering. The two types of preconditioning induced a more uniform genomic response after ischemia/reperfusion. Numerous genes separated preconditioned from unprotected ischemic hearts. Three stable gene clusters were identified in the trigger response to preconditioning, while eight stable clusters related to cytoprotection, inflammation, remodeling, and long interspersed nucleotide elements (LINEs) were delineated after prolonged ischemia. A single stable sample cluster emerged from cluster analysis for both IschPC and unprotected myocardium, indicating a close molecular relationship between these two treatments. Principal component analysis revealed differences between PPC vs. IschPC, and trigger vs. mediator/effector responses in transcripts predominantly related to biosynthesis and apoptosis. IschPC and PPC similarly but distinctly reprogram the genetic response to ischemic injury. IschPC elicits a postischemic gene expression profile closer to unprotected myocardium than PPC, which may be therefore more advantageous as therapeutic strategy in cardioprotection.

Analysis of Variance↗
PubMed2000

Exploring Alternative Models of Complex Patient Management with Artificial Neural Networks.

This study applied an unsupervised neural network modeling process to test data of the National Board of Medical Examiners (NBME) Computer-based Clinical Scenarios (CCS) to identify new performance categories and validate this process as a scoring technique. The classifications resulting from this neural network modeling were consistent with the NBME model in that highly rated NMBE performances (ratings of 7 or 8) were clustered together on the neural network output grid. Very low performance ratings appeared to share few common features and were accordingly classified at isolated nodes. This clustering was reproducible across three separately trained networks with greater than 80% agreement in two of the three networks trained. However, the neural network also contained performance clusters where disparate NBME-based ratings ranged from 1 (worst) to 8 (best). Here, agreement between networks was less than 60%. Through visualization of the search strategies (search path mapping), this neural network clustering was found to be sensitive to quantitative and qualitative test selections such as excessive usage of irrelevant tests reflecting broader behavioral classification in some instances. A disparity between NBME ratings and an independent human rating system was detected by the neural network model since disagreement among raters was also reflected by a lack of neural network performance clustering. Agreement between rating systems, however, was correlated with neural network clustering for 92% of the highly rated performances.

Journal Article↗
PubMed2004

Multispectral classification techniques for terahertz pulsed imaging: an example in histopathology.

Terahertz pulsed imaging is a spectroscopic imaging modality using pulses of electromagnetic radiation (100 GHz-10 THz), and there has been recent interest in studying biomedical specimens. It is usual to display parametric images derived from the measured pulses. In this work, classification was achieved by applying multispectral clustering techniques to sets of parametric images. It was hypothesised that adequate information for clustering was carried in a small number of parametric images, providing these were weighted by complementary physical properties. Materials prepared for histopathological examination were chosen because their condition remained stable during long imaging periods and because their dehydrated state led to greater penetration of the radiation. Two specimens were examined in this pilot study, one of basal cell carcinoma and one of melanoma. Unsupervised ISODATA classification using three selected parametric terahertz pulsed images was compared qualitatively with k-means classification using the shape of the whole time series, and with conventional stained microscope slides. There was good qualitative agreement between the classifications. Classifications were consistent with the morphological appearances expected, but further work is required to determine if tumour discrimination is possible. The results have implications for the future development of the technique as the need for only a small number of features could lead to considerably reduced acquisition times.

Algorithms↗
PubMed2003

Unsupervised feature selection via two-way ordering in gene expression analysis.

MOTIVATION: Selection of genes most relevant and informative for certain phenotypes is an important aspect in gene expression analysis. Most current methods select genes based on known phenotype information. However, certain set of genes may correspond to new phenotypes which are yet unknown, and it is important to develop novel effective selection methods for their discovery without using any prior phenotype information. RESULTS: We propose and study a new method to select relevant genes based on their similarity information only. The method relies on a mechanism for discarding irrelevant genes. A two-way ordering of gene expression data can force irrelevant genes towards the middle in the ordering and thus can be discarded. Mechanisms based on variance and principal component analysis are also studied. When applied to expression profiles of colon cancer and leukemia, the unsupervised method outperforms the baseline algorithm that simply uses all genes, and it also selects relevant genes close to those selected using supervised methods. SUPPLEMENT: More results and software are online: http://www.nersc.gov/~cding/2way.

Algorithms↗
PubMed2004

Dynamic tracking of acute ischemic tissue fates using improved unsupervised ISODATA analysis of high-resolution quantitative perfusion and diffusion data.

High-resolution (200 x 200 x 1,500 microm3) imaging was performed to derive quantitative cerebral blood flow (CBF) and apparent diffusion coefficient (ADC) maps in stroke rats (permanent occlusion) every 30 minutes up to 3 hours after occlusion onset, followed by histology at 24 hours. An improved automated iterative-self-organizing-data-analysis-algorithm (ISODATA) was developed to dynamically track ischemic tissue fate on a pixel-by-pixel basis during the acute phase. ISODATA-resolved clusters were overlaid on the CBF-ADC scatterplots and image spaces. Tissue volume ADC, and CBF of each ISODATA cluster were derived. In contrast to the single-cluster normal left hemisphere (ADC = 0.74 +/- 0.02 x 10(-3) mm2/s, CBF = 1.36 +/- 0.22 mL g(-1)min(-1), mean +/- SD, n = 8), the right ischemic hemisphere exhibited three ISODATA clusters, namely: "normal" (normal ADC and CBF), "ischemic core" (low CBF and ADC), and at-risk "perfusion-diffusion mismatch" (low CBF but normal ADC). At 180 minutes, the mismatch disappeared in five rats (Group I, 180-minute "core" lesion volume = 255 +/- 62 mm3 and 24-hour infarct volume = 253 +/- 55 mm3, P > 0.05), while a substantial mismatch persisted in three rats (Group II, 180-minute CBF-abnormal volume = 198 +/- 7 mm3 and 24-hour infarct volume 148 +/- 18 mm3, P < 0.05). The CBF (0.3 +/- 0.09 mL g(-1)min(-1)) of the "persistent mismatch" (Group II, 0.3 +/- 0.09 mL g(-1)min(-1)) was above the CBF viability threshold (0.2 to 0.3 mL g(-1)min(-1)) throughout and its ADC (0.70 +/- 0.03 x 10(-3) mm2/s) did not decrease as ischemia progressed. In contrast, the CBF (0.08 +/- 0.03 mL g(-1)min(-1)) of the analogous brain region in Group I was below the CBF viability threshold, and its ADC gradually decreased from 0.63 +/- 0.05 to 0.43 +/- 0.03 x 10(-3) mm2/s (ADC viability threshold = 0.53 +/- 0.02 x 10(-3) mm2/s). The modified ISODATA analysis of the ADC and CBF tissue characteristics during the acute phase could provide a useful and unbiased means to characterize and predict tissue fates in ischemic brain injury and to monitor therapeutic intervention.

Algorithms↗
PubMed2004

Joint explorative analysis of neuroreceptor subsystems in the human brain: application to receptor-transporter correlation using PET data.

Positron emission tomography (PET) has proved to be a highly successful technique in the qualitative and quantitative exploration of the human brain's neurotransmitter-receptor systems. In recent years, the number of PET radioligands, targeted to different neuroreceptor systems of the human brain, has increased considerably. This development paves the way for a simultaneous analysis of different receptor systems and subsystems in the same individual. The detailed exploration of the versatility of neuroreceptor systems requires novel technical approaches, capable of operating on huge parametric image datasets. An initial step of such explorative data processing and analysis should be the development of novel exploratory data-mining tools to gain insight into the "structure" of complex multi-individual, multi-receptor data sets. For practical reasons, a possible and feasible starting point of multi-receptor research can be the analysis of the pre- and post-synaptic binding sites of the same neurotransmitter. In the present study, we propose an unsupervised, unbiased data-mining tool for this task and demonstrate its usefulness by using quantitative receptor maps, obtained with positron emission tomography, from five healthy subjects on (pre-synaptic) serotonin transporters (5-HTT or SERT) and (post-synaptic) 5-HT(1A) receptors. Major components of the proposed technique include the projection of the input receptor maps to a feature space, the quasi-clustering and classification of projected data (neighbourhood formation), trans-individual analysis of neighbourhood properties (trajectory analysis), and the back-projection of the results of trajectory analysis to normal space (creation of multi-receptor maps). The resulting multi-receptor maps suggest that complex relationships and tendencies in the relationship between pre- and post-synaptic transporter-receptor systems can be revealed and classified by using this method. As an example, we demonstrate the regional correlation of the serotonin transporter-receptor systems. These parameter-specific multi-receptor maps can usefully guide the researchers in their endeavour to formulate models of multi-receptor interactions and changes in the human brain.

Adult↗
PubMed2001

Basic microarray analysis: grouping and feature reduction.

DNA microarray technologies are useful for addressing a broad range of biological problems - including the measurement of mRNA expression levels in target cells. These studies typically produce large data sets that contain measurements on thousands of genes under hundreds of conditions. There is a critical need to summarize this data and to pick out the important details. The most common activities, therefore, are to group together microarray data and to reduce the number of features. Both of these activities can be done using only the raw microarray data (unsupervised methods) or using external information that provides labels for the microarray data (supervised methods). We briefly review supervised and unsupervised methods for grouping and reducing data in the context of a publicly available suite of tools called CLEAVER, and illustrate their application on a representative data set collected to study lymphoma.

Algorithms↗
PubMed2025

Alzheimer's subtypes A supervised, unsupervised, multimodal, multilayered embedded recursive (SUMMER) AI study.

Since Alzheimer's disease (AD) is a heterogeneous disease, different subtypes may have distinct biological, genetic, and clinical characteristics, requiring tailored interventions. While several proposed subtypes of AD exist, there is still no clear consensus on a definitive classification. By leveraging complementary AI approaches, including supervised and unsupervised learning, within a recursive pipeline (SUMMER) that integrates multimodal datasets encompassing MRI measurements, phenotypes, and genetic data, our goal was to generate robust scientific evidence for identifying AD subtypes. Data was downloaded from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database and included neuroimaging data (MRI), genetics (SNPs), clinical diagnosis, and demographics. 1133 European American participants' images, aged 55-95, were included in this study. The analysis was multi-fold, where the first step involved applying an unsupervised application to a subset of the MRI sample (AD + cognitively normal (CN) aged matched groups, 100 men aged 68-85 years, and 76 women aged 68-85 years). The MRI brain gray matter was segmented into 44 regions of interest (ROIs) according to a standard atlas, and 618 features were extracted, including ROI voxel intensity measurements such as minimum, maximum, and histogram variables. Results identified a cluster of subtype AD men and a cluster of subtype AD women that were distinct from the rest of their respective samples. In the next step, the integrity of the identified subtype AD clusters was investigated using the XGBoost supervised machine learning application with genetic features (SNPs, N=36,724) and labels: the identified subtype AD cluster vs. the rest of the sample, stratified by sex. A significant AD subtype men model (accuracy=0.85, F1=0.72, AUC=0.83) and a significant women AD subtype model (accuracy=0.81, F1=0.81, AUC=0.81) were built, confirming the homogeneity of the isolated AD subtype clusters. Discriminative biomarkers were extracted from the significant models, including selected ROIs and SNPs. Finally, the subtype models were tested on an unseen subset of ADNI data. The genetic-based models identified clusters of AD subtype participants consisting of 34% of the men AD group and 47% of the women AD group. Phenotypic analysis indicates that lower body weight was associated with the women's AD subtype. Complex diseases like AD demand a sophisticated, multimodal approach for precise diagnosis. Effectively identifying disease subtypes enhances the potential for personalized treatment, ultimately improving patient outcomes.

Journal Article↗
PubMed2005

Learning viewpoint invariant object representations using a temporal coherence principle.

Invariant object recognition is arguably one of the major challenges for contemporary machine vision systems. In contrast, the mammalian visual system performs this task virtually effortlessly. How can we exploit our knowledge on the biological system to improve artificial systems? Our understanding of the mammalian early visual system has been augmented by the discovery that general coding principles could explain many aspects of neuronal response properties. How can such schemes be transferred to system level performance? In the present study we train cells on a particular variant of the general principle of temporal coherence, the "stability" objective. These cells are trained on unlabeled real-world images without a teaching signal. We show that after training, the cells form a representation that is largely independent of the viewpoint from which the stimulus is looked at. This finding includes generalization to previously unseen viewpoints. The achieved representation is better suited for view-point invariant object classification than the cells' input patterns. This property to facilitate view-point invariant classification is maintained even if training and classification take place in the presence of an--also unlabeled--distractor object. In summary, here we show that unsupervised learning using a general coding principle facilitates the classification of real-world objects, that are not segmented from the background and undergo complex, non-isomorphic, transformations.

Action Potentials↗
PubMed2005

Optimal filtering for spike sorting of multi-site electrode recordings.

We derive an optimal linear filter, to reduce the distortions of the peak amplitudes of action potentials in extracellular multitrode recordings, which are due to background activity and overlapping spikes. This filter is being learned very efficiently from the raw recordings in an unsupervised manner and responds to the average waveform with an impulse of minimal width. The average waveform does not have to be known in advance, but is learned together with the optimal filter. The peak amplitude of a filtered waveform is a more reliable estimate for the amplitude of an action potential than the peak of the biphasic waveform and can improve the accuracy of the event detection and clustering procedures. We demonstrate a spike-sorting application, in which events are detected using the Mahalanobis distance in the N-dimensional space of filtered recordings as a distance measure, and the event amplitudes of the filtered recordings are clustered to assign events to individual units. This method is fast and robust, and we show its performance by applying it to real tetrode recordings of spontaneous activity in the visual cortex of an anaesthetized cat and to realistic artificial data derived therefrom.

Action Potentials↗
PubMed2006

Gene expression analyses reveal molecular relationships among 20 regions of the human CNS.

Transcriptional profiling was performed to survey the global expression patterns of 20 anatomically distinct sites of the human central nervous system (CNS). Forty-five non-CNS tissues were also profiled to allow for comparative analyses. Using principal component analysis and hierarchical clustering, we were able to show that the expression patterns of the 20 CNS sites profiled were significantly different from all non-CNS tissues and were also similar to one another, indicating an underlying common expression signature. By focusing our analyses on the 20 sites of the CNS, we were able to show that these 20 sites could be segregated into discrete groups with underlying similarities in anatomical structure and, in many cases, functional activity. These findings suggest that gene expression data can help define CNS function at the molecular level. We have identified subsets of genes with the following patterns of expression: (1) across the CNS, suggesting homeostatic/housekeeping function; (2) in subsets of functionally related sites of the CNS identified by our unsupervised learning analyses; and (3) in single sites within the CNS, indicating their participation in distinct site-specific functions. By performing network analyses on these gene sets, we identified many pathways that are upregulated in particular sites of the CNS, some of which were previously described in the literature, validating both our dataset and approach. In summary, we have generated a database of gene expression that can be used to gain valuable insight into the molecular characterization of functions carried out by different sites of the human CNS.

Adult↗
PubMed1993

Potentially functional regions of nucleic acids recognized by a Kohonen's self-organizing map.

Computer recognition of short functional sites on DNA, such as promoter regions or intron-exon boundaries, has recently attracted much interest. In this paper we have focused our attention on the automatic recognition of relevant features of human nucleic acid sequences by means of an unsupervised artificial neural network model. Sixty messenger RNA and 31 genomic DNA sequences were analysed. The results showed that in mRNA, the minimal similarity 60 base pattern was guanine- and cytosine-rich and located in most sequences in a range of 250 bases from either the middle point of the signal peptide coding region or from the start of the coding region. On DNA sequences a region defined by a cluster of minimal similarity patterns was present in many of the analysed genes. This zone may be related to alternative splicing and DNA methylation.

Algorithms↗