SEARCH · PubMed Health

Results for “Predictive Learning Models”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

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

At least 1,009 records · Page 56Linked to original sources

PubMedDate not supplied

Alcohol consumption in university students: the role of reasons for drinking, coping strategies, expectancies, and personality traits.

Despite the popularity of the social learning perspective of alcohol abuse, there have been limited efforts devoted to developing comprehensive models that delineate the roles of the constituent components of this approach. In the present study, we determined whether reasons for drinking, coping strategies, alcohol expectancies, and personality traits predict binge drinking and alcohol consumption levels in university students. Escape drinking was the sole positive direct predictor of binge drinking. Social drinking predicted alcohol consumption and thereby exerted an indirect influence on binge drinking. Alcohol expectancies played a significant role in the model but only by influencing reasons for drinking. Although the use of alcohol and/or drugs to cope predicted alcohol consumption, none of a variety of other coping strategies exerted a significant influence in the model. Stress responsivity-related personality traits played a significant role, primarily via an influence on alcohol expectancies. These findings provide support for the social learning perspective of alcohol abuse and offer further insight into the factors that contribute to the development of risky alcohol consumption patterns.

Adaptation, Psychological↗

PubMed2002

A Bayesian network model for protein fold and remote homologue recognition.

MOTIVATION: The Bayesian network approach is a framework which combines graphical representation and probability theory, which includes, as a special case, hidden Markov models. Hidden Markov models trained on amino acid sequence or secondary structure data alone have been shown to have potential for addressing the problem of protein fold and superfamily classification. RESULTS: This paper describes a novel implementation of a Bayesian network which simultaneously learns amino acid sequence, secondary structure and residue accessibility for proteins of known three-dimensional structure. An awareness of the errors inherent in predicted secondary structure may be incorporated into the model by means of a confusion matrix. Training and validation data have been derived for a number of protein superfamilies from the Structural Classification of Proteins (SCOP) database. Cross validation results using posterior probability classification demonstrate that the Bayesian network performs better in classifying proteins of known structural superfamily than a hidden Markov model trained on amino acid sequences alone.

Amino Acid Sequence↗

PubMed2005

An SVM-based system for predicting protein subnuclear localizations.

BACKGROUND: The large gap between the number of protein sequences in databases and the number of functionally characterized proteins calls for the development of a fast computational tool for the prediction of subnuclear and subcellular localizations generally applicable to protein sequences. The information on localization may reveal the molecular function of novel proteins, in addition to providing insight on the biological pathways in which they function. The bulk of past work has been focused on protein subcellular localizations. Furthermore, no specific tool has been dedicated to prediction at the subnuclear level, despite its high importance. In order to design a suitable predictive system, the extraction of subtle sequence signals that can discriminate among proteins with different subnuclear localizations is the key. RESULTS: New kernel functions used in a support vector machine (SVM) learning model are introduced for the measurement of sequence similarity. The k-peptide vectors are first mapped by a matrix of high-scored pairs of k-peptides which are measured by BLOSUM62 scores. The kernels, measuring the similarity for sequences, are then defined on the mapped vectors. By combining these new encoding methods, a multi-class classification system for the prediction of protein subnuclear localizations is established for the first time. The performance of the system is evaluated with a set of proteins collected in the Nuclear Protein Database (NPD). The overall accuracy of prediction for 6 localizations is about 50% (vs. random prediction 16.7%) for single localization proteins in the leave-one-out cross-validation; and 65% for an independent set of multi-localization proteins. This integrated system can be accessed at http://array.bioengr.uic.edu/subnuclear.htm. CONCLUSION: The integrated system benefits from the combination of predictions from several SVMs based on selected encoding methods. Finally, the predictive power of the system is expected to improve as more proteins with known subnuclear localizations become available.

Algorithms↗

PubMed1996

A clinician-friendly version of the interpersonal circumplex: structural analysis of social behavior (SASB).

Like the original Interpersonal Circumplex (IPC), the Structural Analysis of Social Behavior (SASB) model was developed in the clinic. Different from and more complicated than the IPC, the SASB model nonetheless is parsimonious. It is consistent with Leary's (1957) original goal of bringing objectivity and clarity to the diagnostic process while acknowledging the complexity and variety of human nature. SASB applications extend from diagnosis into the domains of etiology and treatment. Specific advantages that accrue from the SASB model's more complex structure and assessment techniques include the ability to: (a) define both hostile and friendly differentiation, (b) specifically link social learning experiences with self-concept, (c) define several predictive principles on an a priori basis and confirm them in a variety of data sets, (d) assess personality at all 5 of Leary's levels, (e) define normality and pathology in qualitative rather than quantitative terms, (f) compare and contrast self-ratings with observer ratings using the same metric, (g) generate reasonable parallel models for affect and cognitive style that can help account for "comorbidity" between Diagnostic and Statistical Manual of Mental Disorders (4th ed.; American Psychiatric Association, 1994) Axes I and II, (h) make contributions to understanding personality as a hypothetical construct (i.e., to make testable predictions about etiology and specific treatment interventions), (i) dissect complex communications into their underlying components, and (j) accurately characterize a given relationship through a relatively brief sample of behavior.

Humans↗

PubMed1984

Modelling the stages of the identity theory of object-concept development in infancy.

A computational model is presented for the three stages of development of the object concept in infancy identified by Bower and Wishart in their research. The stages are described by sets of PROLOG clauses that interpret object structures representing the perceptual phenomena interpreted by the infants themselves. The infant's changes between developmental stages can be described by differences between the rules modelling each stage. Three experiments are presented and the behaviour of the PROLOG model is described for each stage of development. Motion, rest, and boundedness of objects constitute the theoretical underpinning of the running PROLOG model and are hypothesized as the invariant aspects of perception that explain the behaviour of the infant at each stage of development. A possible explanation for transitions between stages is offered and justified in part by the output of the model, which in turn is used to predict the behavioural outcome of an experiment.

Child Development↗

PubMedDate not supplied

A model of Pavlovian eyelid conditioning based on the synaptic organization of the cerebellum.

We present a model based on the synaptic and cellular organization of the cerebellum to derive a diverse range of phenomena observed in Pavlovian eyelid conditioning. These phenomena are addressed in terms of critical pathways and network properties, as well as the sites and rules for synaptic plasticity. The theory is based on four primary hypotheses: (1) Two cerebellar sites of plasticity are involved in conditioning: (a) bidirectional long-term depression/potentiation at granule cell synapses onto Purkinje cells (gr-->Pkj) in the cerebellar cortex and (b) bidirectional plasticity in the interpositus nucleus that is controlled by inhibitory inputs from Purkinje cells; (2) climbing fiber activity is regulated to an equilibrium level at which the net strength of gr-->Pkj synapses remains constant unless an unexpected unconditioned stimulus (US) is presented or an expected US is omitted; (3) a time-varying representation of the conditioned stimulus (CS) in the cerebellar cortex permits the temporal discrimination required for conditioned response timing; and (4) the ability of a particular segment of the CS to be represented consistently across trials varies as a function of time since CS onset. This variation in across-trials consistency is thought to contribute to the ISI function. The model suggests several empirically testable predictions, some of which have been tested recently.

Animals↗

PubMed2003

Telephone screening for amnestic mild cognitive impairment.

OBJECTIVES: To evaluate the utility of telephone screening for identifying subjects with amnestic mild cognitive impairment (aMCI) for enrollment in a clinical trial and to identify which elements of the modified Telephone Interview for Cognitive Status (TICS-m) best predicted the in-clinic determination of aMCI. METHODS: Subjects aged >/=65 years with memory complaints responded to an advertisement for a clinical trial by calling a central telephone recruiting agency. To determine eligibility, subjects went through a stepwise selection procedure involving a review of major protocol inclusion and exclusion criteria, followed by administration of the Category Fluency Test (CFT) and then the TICS-m. Subjects meeting entry criteria, who obtained a score of </=13 on the CFT for "animals" and </=24 on the CFT for "animals" and "fruits" and who scored between 19 and 38 on the TICS-m, were referred for a clinic appointment to determine whether they met clinical criteria for aMCI. Clinical criteria for aMCI required a score of >/=24 on the Mini-Mental State Examination and a score of </=37 on the Rey Auditory Verbal Learning Test. A post hoc analysis was performed using factor analysis and logistic regression models to investigate which elements of the TICS-m best predicted the in-clinic determination of aMCI. RESULTS: Of 16,988 subjects who called the telephone agency, 8,742 passed the review of inclusion/exclusion criteria; 6,090 met the CFT cut scores and received the TICS-m; 5,223 met cut scores on the TICS-m and were referred for an in-clinic appointment; 747 were seen in the clinic; and 324 met clinical criteria for aMCI. Factor analysis indicated three factors on the TICS-m: language/attention, orientation, and memory. The memory factor, comprising immediate and delayed recall of a word list, was the most important contributor for identifying subjects who met clinical criteria for aMCI. CONCLUSION: Only 2% of subjects who underwent telephone screening were recruited into the study, but 43% of those who passed telephone screening and were seen in the clinic met clinical criteria for aMCI. The word recall tests of the TICS-m were the most important items for identifying which subjects met clinical criteria for aMCI.

Age Distribution↗

PubMed2005

Identifying simple discriminatory gene vectors with an information theory approach.

In the feature selection of cancer classification problems, many existing methods consider genes individually by choosing the top genes which have the most significant signal-to-noise statistic or correlation coefficient. However the information of the class distinction provided by such genes may overlap intensively, since their gene expression patterns are similar. The redundancy of including many genes with similar gene expression patterns results in highly complex classifiers. According to the principle of Occam's razor, simple models are preferable to complex ones, if they can produce comparable prediction performances to the complex ones. In this paper, we introduce a new method to learn accurate and low-complexity classifiers from gene expression profiles. In our method, we use mutual information to measure the relation between a set of genes, called gene vectors, and the class attribute of the samples. The gene vectors are in higher-dimensional spaces than individual genes, therefore, they are more diverse, or contain more information than individual genes. Hence, gene vectors are more preferable to individual genes in describing the class distinctions between samples since they contain more information about the class attribute. We validate our method on 3 gene expression profiles. By comparing our results with those from literature and other well-known classification methods, our method demonstrated better or comparable prediction performances to the existing methods, however, with lower-complexity models than existing methods.

Algorithms↗

PubMed2005

Knowledge discovery and system biology in molecular medicine: an application on neurodegenerative diseases.

The possibility to study an organism in terms of system theory has been proposed in the past, but only the advancement of molecular biology techniques allow us to investigate the dynamical properties of a biological system in a more quantitative and rational way than before . These new techniques can gave only the basic level view of an organisms functionality. The comprehension of its dynamical behaviour depends on the possibility to perform a multiple level analysis. Functional genomics has stimulated the interest in the investigation the dynamical behaviour of an organism as a whole. These activities are commonly known as System Biology, and its interests ranges from molecules to organs. One of the more promising applications is the 'disease modeling'. The use of experimental models is a common procedure in pharmacological and clinical researches; today this approach is supported by 'in silico' predictive methods. This investigation can be improved by a combination of experimental and computational tools. The Machine Learning (ML) tools are able to process different heterogeneous data sources, taking into account this peculiarity, they could be fruitfully applied to support a multilevel data processing (molecular, cellular and morphological) that is the prerequisite for the formal model design; these techniques can allow us to extract the knowledge for mathematical model development. The aim of our work is the development and implementation of a system that combines ML and dynamical models simulations. The program is addressed to the virtual analysis of the pathways involved in neurodegenerative diseases. These pathologies are multifactorial diseases and the relevance of the different factors has not yet been well elucidated. This is a very complex task; in order to test the integrative approach our program has been limited to the analysis of the effects of a specific protein, the Cyclin dependent kinase 5 (CDK5) which relies on the induction of neuronal apoptosis. The system has a modular structure centred on a textual knowledge discovery approach. The text mining is the only way to enhance the capability to extract ,from multiple data sources, the information required for the dynamical simulator. The user may access the publically available modules through the following site: http://biocomp.ge.ismac.cnr.it.

Biomedical Research↗

PubMed2004

Operon prediction without a training set.

MOTIVATION: Annotation of operons in a bacterial genome is an important step in determining an organism's transcriptional regulatory program. While extensive studies of operon structure have been carried out in a few species such as Escherichia coli, fewer resources exist to inform operon prediction in newly sequenced genomes. In particular, many extant operon finders require a large body of training examples to learn the properties of operons in the target organism. For newly sequenced genomes, such examples are generally not available; moreover, a model of operons trained on one species may not reflect the properties of other, distantly related organisms. We encountered these issues in the course of predicting operons in the genome of Bacteroides thetaiotaomicron (B.theta), a common anaerobe that is a prominent component of the normal adult human intestinal microbial community. RESULTS: We describe an operon predictor designed to work without extensive training data. We rely on a small set of a priori assumptions about the properties of the genome being annotated that permit estimation of the probability that two adjacent genes lie in a common operon. Predictions integrate several sources of information, including intergenic distance, common functional annotation and a novel formulation of conserved gene order. We validate our predictor both on the known operons of E.coli and on the genome of B.theta, using expression data to evaluate our predictions in the latter.

Algorithms↗

PubMed1998

[Anticipatory nausea and anticipatory vomiting, food aversion and anticipatory immunomodulation: classical conditioning in cytostatic drug treatment of pediatric chemotherapy patients].

Nausea and/or vomiting are adverse side-effects of cancer chemotherapeutic drugs in adult as well as pediatric cancer patients' complicating treatment and compliance. Nausea and vomiting are not only experienced as posttreatment symptoms after chemotherapy (i.e., posttreatment nausea and/or vomiting). In a subgroup of cancer patients, these symptoms also occur prior to a chemotherapeutic drug infusion, called anticipatory nausea (AN) and anticipatory vomiting (AV). The aim of this paper is to present a model derived from basic psychology to explain anticipatory symptoms as learned responses based on classical conditioning. In addition, food aversions and also immunomodulation are interpreted as conditioned responses. Some data on prevalence of ANV in a pediatric sample and on the correspondence between anticipatory symptoms and predictions from the conditioning model are presented. Finally, therapeutic techniques to prevent AN and/or AV are deduced from the conditioning model.

Adult↗

PubMed1992

Transmission of the fra(X) haplotype from three nonpenetrant brothers to their affected grandsons.

We report on a family showing transmission of the fra(X) gene by 3 nonpenetrant, fra(X) negative, normally intelligent, full and half-brothers to their affected grandsons. The mothers of the affected boys are obligate carriers, fra(X) negative, and of normal intelligence. This family illustrates the "Sherman Paradox" and is compatible with the predictions of the Laird X-inactivation imprinting model. In addition, molecular and/or cytogenetic studies have enabled at-risk relatives to learn more about their carrier fra(X) status and have allowed for more accurate genetic counselling.

Chromosome Mapping↗

PubMed2006

Reconstructing protein complexes: from proteomics to systems biology.

Modern high throughput technologies in biological science often create lists of interesting molecules. The challenge is to reconstruct a descriptive model from these lists that reflects the underlying biological processes as accurately as possible. Once we have such a model or network, what can we learn from it? Specifically, given that we are interested in some biological process associated with the model, what new properties can we predict and subsequently test? Here, we describe, at an introductory level, a range of bioinformatics techniques that can be systematically applied to proteomic datasets. When combined, these methods give us a global overview of the network and the properties of the proteins and their interactions. These properties can then be used to predict functional pathways within the network and to examine substructure. To illustrate the application of these methods, we draw upon our own work concerning a complex of 186 proteins found in neuronal synapses in mammals. The techniques discussed are generally applicable and could be used to examine lists of proteins involved with the biological response to electric or magnetic fields.

Animals↗

PubMed2005

Interactome modeling.

A long-term goal of the field of interactome modeling is to understand how global and local properties of complex macromolecular networks impact on observable biological properties, and how changes in such properties can lead to human diseases. The information available at this stage of development of the field provides strong evidence for the existence of such interesting global and local properties, but also demonstrates that many more datasets will be needed to provide accurate models with increasingly predictive capacity. This review focuses on an early attempt at mapping a multicellular interactome network and on the lessons learned from that attempt.

Animals↗

PubMed2005

Poor premorbid social functioning and theory of mind deficit in schizophrenia: evidence of reduced context processing?

Investigations have demonstrated deficits in theory of mind (ToM) ability in schizophrenia. Yet, the development of, and mechanisms associated with these deficits are not well understood. The present investigation examined the hypothesis that, among chronic schizophrenia patients, impaired ToM is associated with failures in context processing, greater disorganized symptoms, and poor premorbid functioning. Forty-two inpatients with schizophrenia spectrum disorders were assessed on tests of ToM, visual and linguistic context processing, executive functioning, and verbal IQ. Symptomatology and premorbid functioning were also assessed. Results revealed that more impaired ToM was associated with poorer performance on both visual and linguistic context processing measures and higher ratings of disorganization on the BRRS. ToM was also associated with poorer childhood social functioning and an earlier age of illness onset. ToM was not associated with verbal processing speed, verbal fluency, response inhibition, sequence learning, or estimated verbal IQ. A significant regression model including measures of childhood peer problems and visual and language context processing significantly predicted ToM performance and accounted for 43% of the variance. These findings suggest that, among chronic schizophrenia patients, deficits in ToM ability may be the result of context processing impairments. These impairments may be a factor in both poor social functioning during childhood and greater disorganized symptoms after illness onset.

Adult↗

PubMed2006

Post-training amphetamine administration enhances memory consolidation in appetitive Pavlovian conditioning: Implications for drug addiction.

It has been suggested that some of the addictive potential of psychostimulant drugs of abuse such as amphetamine may result from their ability to enhance memory for drug-related experiences through actions on memory consolidation. This experiment examined whether amphetamine can specifically enhance consolidation of memory for a Pavlovian association between a neutral conditioned stimulus (CS-a light) and a rewarding unconditioned stimulus (US-food), as Pavlovian conditioning of this sort plays a major role in drug addiction. Male Long-Evans rats were given six training sessions consisting of 8 CS presentations followed by delivery of the food into a recessed food cup. After the 1st, 3rd, and 5th session, rats received subcutaneous injections of amphetamine (1.0 or 2.0 mg/kg) or saline vehicle immediately following training. Conditioned responding was assessed using the percentage of time rats spent in the food cup during the CS relative to a pre-CS baseline period. Both amphetamine-treated groups showed significantly more selective conditioned responding than saline controls. In a control experiment, there were no differences among groups given saline, 1.0 or 2.0 mg/kg amphetamine 2 h post-training, suggesting that immediate post-training amphetamine enhanced performance specifically through actions on memory consolidation rather than through non-mnemonic processes. This procedure modeled Pavlovian learning involved in drug addiction, in which the emotional valence of a drug reward is transferred to neutral drug-predictive stimuli such as drug paraphernalia. These data suggest that amphetamine may contribute to its addictive potential through actions specifically on memory consolidation.

Amphetamine↗

PubMed1999

Anti-DNA autoantibodies and systemic lupus erythematosus.

Systemic lupus erythematosus (SLE) is a systemic autoimmune disease that affects most of the organs and tissues of the body, causing glomerulonephritis, arthritis, and cerebritis. SLE can be fatal with nephritis, in particular, predicting a poor outcome for patients. In this review, we highlight what has been learned about SLE from the study of mouse models, and pay particular attention to anti-DNA autoantibodies, both as pathological agents of lupus nephritis and as DNA-binding proteins. We summarize the current approaches used to treat SLE and discuss the targeting of anti-DNA autoantibodies as a new treatment for lupus nephritis.

Animals↗

PubMed2000

Representing the task in Bayesian reasoning: comment on Lovett and Schunn (1999).

The RCCL model (M. C. Lovett & C. D. Schunn, 1999) produces predictions that are non-novel or that do not truly spring from its principles. However, it offers the valuable insight that learning processes may affect the selection of both representations and strategies within those representations, and points the way to possible theoretical progress on implicit and explicit control. The authors' account of base-rate neglect under direct experience is compared with RCCL, and it is concluded that learning-based models allow for tests that are not fostered by representation-based models.

Bayes Theorem↗