SEARCH · PubMed Health

Results for “Large Language 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 415 records · Page 23Linked to original sources

PubMed1995

Familial sinistrality: the utility of calculating exact genotype probabilities for individuals.

Measures of familial sinistrality are frequently used in neuropsychological research. However they suffer from the problem that they are essentially a global measure of phenotypes in the family, whereas they are interpreted as an indicator of the genotype of the individual concerned. In this paper it is shown how to calculate a precise probabilistic estimate of a proband's genotype, given a particular genetic model of handedness, using all of the information available in a family tree. Example calculations are provided for a range of families with one or two sinistral relatives. Genotype probabilities are also calculated for a large population data set (the National Childhood Encephalopathy Study) and it is shown that familial sinistrality does not account for all of the variability present in the genotype probabilities, and that genotype probabilities provide an additional prediction of individual sinistrality after taking familial sinistrality into account, whereas all of the information in familial sinistrality is accounted for by the genotype probabilities. Finally it is shown how genotype probabilities can be used to assess whether there is heterozygote advantage for a characteristic, and using this method it is suggested that there is no support for heterozygotes having higher intellectual ability than homozygotes.

Family↗

PubMed2005

Competitive hybridization kinetics reveals unexpected behavior patterns.

Although the kinetics of hybridization between a soluble polynucleotide and an immobilized complementary sequence have been studied by others, it is almost universally assumed that the interaction between each probe/target pair can be treated as a separate event. This simplifies the mathematics considerably, but it can give a false picture of the extent of hybridization that one achieves at equilibrium as well as the relative quantities of each hybridized pair during the approach to equilibrium. Here we solve the relevant kinetics equations simultaneously using Mathematica as a simulation language. Among the interesting results of this study are that, for certain circumstances, the relative ratio of incorrect to correct hybrids can change dramatically with time; that the relative abundances of two pairs are not what one would expect based on their equilibrium dissociation constants; that the volume of a wash solution after hybridization can have a large effect on results; and the fact that a short wash is typically better than a long one. We show that an optimum wash time exists for a given set of conditions. In addition, the ratio of soluble to insoluble (spotted) molecules can influence results substantially. Finally, the true levels of rare transcripts can be masked by the presence of highly abundant ones. Code is supplied to enable others to study conditions beyond those presented in this article.

Base Composition↗

PubMed1979

A hospital medication monitoring system.

Hospitals need to process large quantities of scattered information to properly screen medication orders. The data includes complex relationships involving drugs, diseases, and other drugs that determine whether a particular drug is contraindicated for a patient. Hospitals would therefore benefit from a flexible system for storing complex medical data relationships. We have used GPLAN, a CODASYL-like data management system, to construct a functioning model of a hospital pharmacy medication monitoring system. the patients' histories, drug-drug contraindications, disease-drug contraindications, and recommended dosages are stored in the data base in a network structure defined by the Data Description Language. Each drug order is checked for dose, allergies, current diseases, and current medications.

Computers↗

PubMed1995

Working memory networks and the origin of language areas in the human brain.

Temporoparietal-prefrontal working memory networks are proposed as fundamental in the evolutionary origin of the language regions. Having a primordial capacity to name objects or situations, primitive hominids may have strongly benefited from the possibility to recall past events from memory, in order to refer to them through vocal communication. Working memory cortical networks are related to these types of tasks, and are arranged quite similarly to the language networks in the brain. It is possible that the language areas and their connections arose as a local specialization of these large-scale cortical networks, that developed as neural strategies to recall past events to be shared in community. The origin of syntax may have taken place after these networks were sufficiently stabilized, and (at least originally) may have been related to aspects of vocal motor control, involving the progressive differentiation of the anterior language areas and their connections.

Biological Evolution↗

PubMedDate not supplied

The Internet and medical collaboration using virtual reality.

Computed Tomography (CT) provides a large amount of data but the presentation of the data to a physician can be less than satisfactory. Ideally, the image data should be available to physicians in interactive 3D to allow for improved visualization, planning and diagnosis. A virtual reality representation that not only allows for the manipulation of the image but also allows for the user to, in effect, move inside the image remotely would be ideal. In this paper the research associated with virtual reality is discussed. A formalism is then presented to create, from the CT data, the virtual reality world in the Virtual Reality Modeling Language. An implementation is described of this formalism that uses the Internet to allow for users in remote locations to view and manipulate the virtual worlds.

Internet↗

PubMed1987

[An expert system of aiding decision making in breast pathology connected to a clinical data base].

The René Huguenin Cancer Center holds a medical file for each patient which is intended to store and process medical data. Since 1970, we introduced computerization: a development plan was elaborated and simultaneously a statistical software (Clotilde--GSI/CFRO) was selected. Thus, we now have access to a large database, structured according to medical rationale, and utilizable with methods of artificial intelligence towards three objectives: improved data acquisition, decision making and exploitation. The first application was to breast pathology, which represents one of the Center's primary activities. The structure of the data concerning patients is by all criteria part of the medical knowledge. This information needs to be presented as well as processed with a suitable language. To this end, we chose a language-oriented object, Mering II, usable with Apple and IBM 4 micro-computers. This project has already allowed to work out an operational model.

Breast Neoplasms↗

PubMed1999

Paradigms and personhood: a deepening of the dilemmas in ethics and medical ethics.

There are many calls for a definition of personhood, but also many logical and Wittgensteinian reasons to think fulfilling this is unimportant or impossible. I argue that we can consider many contexts as language-games and consider the person as the key player in each. We can then examine the attributes, presuppositions and implications of personhood in those contexts. I use law and therapeutic psychology as two examples of such contexts or language-games. Each correlates with one of the classic "theories" of ethics-deontology and consequentialism. But each is a large enough cluster to consider them as paradigms in a sense related to Thomas Kuhn's notion in The Structure of Scientific Revolutions. Showing the presuppositions about and "takes" on personhood together with the connections involved in the paradigms deepens the dilemmas we already know to be present.

Bioethical Issues↗

PubMed1988

Dynamic pattern generation in behavioral and neural systems.

In the search for principles of pattern generation in complex biological systems, an operational approach is presented that embraces both theory and experiment. The central mathematical concepts of self-organization in nonequilibrium systems (including order parameter dynamics, stability, fluctuations, and time scales) are used to show how a large number of empirically observed features of temporal patterns can be mapped onto simple low-dimensional (stochastic, nonlinear) dynamical laws that are derivable from lower levels of description. The theoretical framework provides a language and a strategy, accompanied by new observables, that may afford an understanding of dynamic patterns at several scales of analysis (including behavioral patterns, neural networks, and individual neurons) and the linkage among them.

Animals↗

PubMed2000

Serial control of phonology in speech production: a hierarchical model.

A dynamic oscillator-based model of the sequencing of phonemes in speech production (OSCAR) is described. An analysis of phoneme movement errors (anticipations, perseverations, and exchanges) from a large naturalistic speech error corpus provides a new set of data suitable for quantitative modeling and is used to derive a set of constraints that any speech-production model must address. The new computational model is shown to account for error type proportions, movement error distance gradients, the syllable-position effect, and phonological similarity effects. The model provides an alternative to frame-based accounts, serial buffer accounts, and associative chaining theories of serial order processing in speech.

Humans↗

PubMed2003

Language dynamics in finite populations.

Any mechanism of language acquisition can only learn a restricted set of grammars. The human brain contains a mechanism for language acquisition which can learn a restricted set of grammars. The theory of this restricted set is universal grammar (UG). UG has to be sufficiently specific to induce linguistic coherence in a population. This phenomenon is known as "coherence threshold". Previously, we have calculated the coherence threshold for deterministic dynamics and infinitely large populations. Here, we extend the framework to stochastic processes and finite populations. If there is selection for communicative function (selective language dynamics), then the analytic results for infinite populations are excellent approximations for finite populations; as expected, finite populations need a slightly higher accuracy of language acquisition to maintain coherence. If there is no selection for communicative function (neutral language dynamics), then linguistic coherence is only possible for finite populations.

Humans↗

PubMed1994

Functional pronunciation units in English words.

Two distinct factors limit the orthographic regularity of English words: (a) Most characters can correspond to several different sounds and (b) characters can either stand alone or be combined in various ways for pronunciation as a single phoneme. This study addresses the second of these issues through the analysis of a large corpus of English words. Data are presented describing the frequency that each character (or character cluster) functioned in the corpus as a correspondent of a single phoneme rather than being combined with other characters (or decomposed). Examples are provided regarding potential applications of these data in the construction of stimulus materials for cognitive studies, in neuropsychological investigations of dyslexia, and in computational models of word naming.

Humans↗

PubMed1981

[Ventilation perfusion distribution: a program for personal computer (author's transl)].

Ventilation perfusion inequality, is analysed with the help of a program written in Basic language, easily implemented on a personal computer. For a gas exchange unit, the main equation describes a mass balance for oxygen, assuming absence of diffusion barrier. The use of a large number of these units allows for calculation of the resulting arterial PO2 in case of abnormal ventilation blood flow distribution. Several simulations are made to study the effect of local variations of VA/Q ratio, and oxygen inhalation upon arterial PO2.

Computers↗

PubMed2003

Unfolding of phonetic information over time: a database of Dutch diphone perception.

We present the results of a large-scale study on speech perception, assessing the number and type of perceptual hypotheses which listeners entertain about possible phoneme sequences in their language. Dutch listeners were asked to identify gated fragments of all 1179 diphones of Dutch, providing a total of 488,520 phoneme categorizations. The results manifest orderly uptake of acoustic information in the signal. Differences across phonemes in the rate at which fully correct recognition was achieved arose as a result of whether or not potential confusions could occur with other phonemes of the language (long with short vowels, affricates with their initial components, etc.). These data can be used to improve models of how acoustic-phonetic information is mapped onto the mental lexicon during speech comprehension.

Adult↗

PubMedDate not supplied

[Genetic studies in communication disorders].

OBJECTIVES: To review current literature on population, cytogenetic and molecular studies of specific language disorders (SLD) and pervasive developmental disorders (PDD). DEVELOPMENT: Clinical concordance studies in twins and in vertical familial groups suggest polygenic multifactorial modes of inheritance, but in some families an autosomal dominant model may be present. The data favour not a modular, but rather a molar model of the relationship between genes and neural abilities for communicative behaviors. Several extensive genome screenings have demonstrated linkage to specific markers on 7q for SLD, and on 7q and 2q for PDD. The strong evidence of linkage on 7q for both disorders has led to the hypothesis that this region contains several separate quantitative trait loci (QTL) related to different communicative abilities. Mutations in different QTL would facilitate the different disabilities and stereotyped behaviors associated with the phenotypic spectrum of PDD. There are other candidate regions for QTLs but the linkage is weaker and there is little agreement between studies; due, in part, to over extensive inclusion criteria and small sizes of familial groups. CONCLUSIONS: To enhance linkage research in further molecular genetic studies, clinicians must refine behavioral target traits when selecting familial groups and enlarge the size of familial groups by including non handicapped members with related behavioral traits. At present, a chromosome region in 7q shows the strongest evidence for communication related QTL, but other QTL need to be identified elsewhere in the genome in order to explain the genetic contribution to the large spectrum of language and autistic disorders.

Autistic Disorder↗

PubMed2001

Language and number: a bilingual training study.

Three experiments investigated the role of a specific language in human representations of number. Russian-English bilingual college students were taught new numerical operations (Experiment 1), new arithmetic equations (Experiments 1 and 2), or new geographical or historical facts involving numerical or non-numerical information (Experiment 3). After learning a set of items in each of their two languages, subjects were tested for knowledge of those items, and new items, in both languages. In all the studies, subjects retrieved information about exact numbers more effectively in the language of training, and they solved trained problems more effectively than untrained problems. In contrast, subjects retrieved information about approximate numbers and non-numerical facts with equal efficiency in their two languages, and their training on approximate number facts generalized to new facts of the same type. These findings suggest that a specific, natural language contributes to the representation of large, exact numbers but not to the approximate number representations that humans share with other mammals. Language appears to play a role in learning about exact numbers in a variety of contexts, a finding with implications for practice in bilingual education. The findings prompt more general speculations about the role of language in the development of specifically human cognitive abilities.

Adult↗

PubMed2004

CLANS: a Java application for visualizing protein families based on pairwise similarity.

SUMMARY: The main source of hypotheses on the structure and function of new proteins is their homology to proteins with known properties. Homologous relationships are typically established through sequence similarity searches, multiple alignments and phylogenetic reconstruction. In cases where the number of potential relationships is large, for example in P-loop NTPases with many thousands of members, alignments and phylogenies become computationally demanding, accumulate errors and lose resolution. In search of a better way to analyze relationships in large sequence datasets we have developed a Java application, CLANS (CLuster ANalysis of Sequences), which uses a version of the Fruchterman-Reingold graph layout algorithm to visualize pairwise sequence similarities in either two-dimensional or three-dimensional space. AVAILABILITY: CLANS can be downloaded at http://protevo.eb.tuebingen.mpg.de/download.

Algorithms↗

PubMed2026

CLASPP: A unified model for predicting post-translational modifications.

Post-Translational Modifications (PTMs) are a fundamental mechanism for regulating cellular pathways and increasing the functional diversity of the proteome. Accurately predicting the PTM types that are likely to occur at a given site in the primary sequence is a key challenge in functional proteomics. Existing PTM prediction models predominantly focus on either single PTM types or employ ensemble methods that combine multiple models to predict different PTM types. This fragmentation is largely driven by the vast imbalance in data availability across PTM types, making it difficult to predict multiple PTM types with a single model. To address this limitation, we present the Contrastively Learned Attention-based Stratified PTM Predictor (CLASPP), a unified PTM prediction model. CLASPP addresses imbalance challenges by leveraging unsupervised clustering-based undersampling and a novel contrastive learning framework tailored to PTM data. Additionally, our hierarchical data organization and curation are shown to improve CLASPP's performance by balancing the representation of individual PTM types and provides a standardized dataset to train and validate future model designs. Drawing inspiration from advancements in image and natural language processing, the CLASPP model employs a multi-stage training strategy and a high-quality, curated training dataset to improve PTM prediction performance. To uncover what is learned during the contrastive learning stage, the CLASPP model is shown to distinguish known protein kinase substrate specificity profiles as a form of explainability. Finally, we evaluate the application of CLASPP in predicting PTMs in different model organisms and experimentally validated ubiquitination sites in the understudied DCLK3 kinase. Overall, CLASPP represents a unified model for PTM prediction that addresses key bottlenecks in data imbalance and offers new strategies for biological data curation, thereby improving PTM-type prediction performance across diverse organisms.

Protein Processing, Post-Translational↗

PubMed1999

Functional differentiation of medial temporal and frontal regions involved in processing novel and familiar words: an fMRI study.

Results of recent functional magnetic resonance imaging (fMRI) studies of memory are not entirely consistent with lesion studies. Furthermore, although imaging probes have identified neural systems associated with processing novel visual episodic information, auditory verbal memory using a novel/familiar paradigm has not yet been examined. To address this gap, fMRI was used to compare the haemodynamic response when listening to recently learned and novel words. Sixteen healthy adults (6 male, 10 female) learned a 10-item word list to 100% criterion, approximately 1 h before functional scanning. During echo-planar imaging, subjects passively listened to a string of words presented at 6-s intervals. Previously learned words were interspersed pseudo-randomly between novel words. Mean scans corresponding to each word type were analysed with a random-effects model using statistical parametric mapping (SPM96). Familiar (learned) words activated the right prefrontal cortex, posterior left parahippocampal gyrus, left medial parietal cortex and right superior temporal gyrus. Novel words activated the anterior left hippocampal region. The results for the familiar words were similar to those found in other functional imaging studies of recognition and retrieval and implicate the right dorsolateral prefrontal and left posterior medial temporal lobe (MTL) regions. The results for novel words require replication, but are consistent with the substantial lesion and PET literature implicating the anterior MTL as a critical site for processing novel episodic information, presumably to permit encoding. Together, these results provide evidence for an anterior-posterior functional differentiation within the MTL in processing novel and familiar verbal information. The differentiation of MTL functions that was obtained is consistent with a large body of PET activation studies but is unique among fMRI studies, which to date have differed from results with PET. Further, the finding of left MTL lateralization is consistent with lesion-based material-specific models of memory.

Adult↗