Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “representation learning”

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 73 records · Page 4Linked to original sources
PubMed2000

Medical text representations for inductive learning.

Inductive learning algorithms have been proposed as methods for classifying medical text reports. Many of these proposed techniques differ in the way the text is represented for use by the learning algorithms. Slight differences can occur between representations that may be chosen arbitrarily, but such differences can significantly affect classification algorithm performance. We examined 8 different data representation techniques used for medical text, and evaluated their use with standard machine learning algorithms. We measured the loss of classification-relevant information due to each representation. Representations that captured status information explicitly resulted in significantly better performance. Algorithm performance was dependent on subtle differences in data representation.

Algorithms↗
PubMed2026

Toward AI Virtual Cells for Hepatology: Representation, Generation, Dynamics, and Intervention in Single-Cell Models.

``Single-cell and spatial atlases describe the healthy and diseased liver at high resolution, including lobular hepatocyte zonation, fibrotic macrophage-stellate niches, cholangiocyte reactions, immune remodeling, and hepatocellular carcinoma ecosystems. These maps show where cell states occur but do not, by themselves, predict whether liver injury will progress or how the liver will respond to an untested drug, toxicant, or genetic perturbation. In this review, we organize current approaches toward an AI Virtual Cell (AIVC) for the liver into three complementary modeling routes. Generative models represent cell states, dynamics and transport models infer state transitions, and pretrained or foundation models test whether learned representations transfer across donors, etiologies, disease stages, and platforms. Perturbation-response prediction serves as a cross-cutting assessment of whether these layers can predict responses to untested genetic, chemical, inflammatory, or metabolic interventions. Available evidence can be categorized as direct liver validation, liver-included benchmarks, general single-cell evidence, and conceptual applications. Published models demonstrate individual components, including atlas integration, inferred trajectories, transferable representations, and retrospective response programs. However, these models do not constitute a prospectively validated liver simulator. At minimum, evaluation should include donor-, etiology-, stage-, platform-, and perturbation-level hold-outs. Model performance should be reported using response direction, recovery of differentially expressed genes and rare states, and calibrated uncertainty. Claims about tissue- or function-level prediction additionally require independent spatial, histologic, metabolic, and functional readouts. Near-term use should prioritize experiment selection and hypothesis generation, whereas clinical decision support remains a longer-term objective.

AI Virtual Cell↗
PubMed1995

What does the hippocampus really do?

Much of the evidence used to implicate the hippocampus in learning and memory has been obtained from clinical cases and/or experimental studies with animals where the damage is extensive and includes more than just the hippocampus. When the damage is limited to the cells that comprise the hippocampus (CA1-CA3 pyramidal cells, hilar and granule cells in the dentate gyrus) the effect on behavior in the rat is more limited than what is usually reported. Selective, axon-sparing ibotenic acid lesions of the hippocampus were used in the experiments that are reviewed to study the effects of removing the hippocampus on: (1) the acquisition of spatial and non-spatial information; (2) complex, non-spatial representational learning; and (3) acquisition and utilization of contextual information. The results indicated that rats with the hippocampus removed were impaired on those tasks that require the utilization of spatial and contextual information but performed like controls in learning about and handling (even complex) non-spatial information. Future research utilizing selective lesions of the hippocampus and sensitive behavioral testing techniques should help clarify the extent to which the impairments in the acquisition of spatial information and the ability to utilize contextual, background cues can be reduced to a single, underlying learning process.

Animals↗
PubMed2026

Biological Foundation Models for Complex Disease Research and Clinical Translation.

Complex diseases, including cancer, rare genetic disorders, neurodevelopmental and psychiatric conditions, and neurodegenerative diseases, arise from interactions among genetic variation, gene regulation, and cellular states that are difficult to capture using a single data type or biological scale. Biological foundation models address this challenge by treating nucleotides and genes as tokens and learning representations that can be transferred to downstream biomedical and clinical tasks. In this review, we examine two major model classes, genomic sequence foundation models and cell foundation models, and compare their tokenization strategies, model architectures, pretraining objectives, and adaptation methods. We summarize their emerging applications in regulatory variant interpretation, disease-associated cell-state analysis, drug-response prediction, and therapeutic target discovery across complex diseases. We distinguish applications supported by experimental or retrospective validation from those that remain primarily computational or conceptual. We further discuss key challenges to clinical translation, including multimodal data integration, model interpretability, benchmarking, patient-specific prediction, and privacy protection. We highlight future opportunities to integrate biological foundation models with emerging frameworks of medical digital twins, agentic AI, and federated learning. By linking model design to translational goals, this review provides a practical framework for evaluating biological foundation models and their readiness for complex disease research and clinical use.

biological foundation model↗
PubMed1988

Crossed buccofacial apraxia.

The cerebral hemisphere contralateral to the preferred hand is generally dominant for learned representational motor acts, including those involving buccofacial muscles. It is generally also language-dominant. This buccofacial apraxia has, with rare exceptions, been associated with left hemispheric lesions in right-handers. We describe two patients with severe buccofacial apraxia caused by large middle cerebral artery territory infarcts in the hemisphere ipsilateral to the preferred hand and nondominant for language. Neither patient had aphasia or major limb apraxia. Computed tomographic scans in the first patient and neuropathologic examination in the second failed to reveal an abnormality of the hemisphere contralateral to the preferred hand. Hence, in some individuals, the hemisphere controlling skilled representational buccofacial movements may not be the one that is dominant either for handedness or for language.

Aged↗
PubMed2026

Atlas-level single-cell integration and clustering-free differential expression analysis with GEDI 2.0.

MOTIVATION: GEDI is a generative framework for multi-sample, multi-condition single-cell analysis that performs batch correction, latent representation learning, and clustering-free differential expression within a unified model. However, the original implementation suffered from prohibitive memory use and runtime, preventing its application to modern atlas-scale datasets. RESULTS: We present GEDI 2.0, a complete high-performance reimplementation featuring a standalone C++ computational core with pre-allocated workspaces, strict sparse-matrix preservation, optimized BLAS routines, and multi-threaded block-coordinate descent. Across extensive benchmarks spanning up to 500 000 cells and 10 000 features, GEDI 2.0 achieves 40%-63.6% mean reduction in peak memory, 2.98× mean single-threaded speedups, and up to 11.5× acceleration with parallel execution, while maintaining full numerical equivalence to the original method. These improvements enable GEDI 2.0 to analyze million-cell datasets, a scale not achievable with the legacy implementation. GEDI 2.0 provides R and Python interfaces and seamless interoperability with common single-cell workflows. AVAILABILITY AND IMPLEMENTATION: Source code, documentation, reproducible codebase, and tutorials are available at https://github.com/csglab/gedi2.

Single-Cell Analysis↗
PubMed2026

ARISE: RNA-anchored shared-edge topology and hierarchical fusion for spatial multi-omics integration.

MOTIVATION: Spatial multi-omics technologies jointly profile transcriptomes, proteins and chromatin accessibility in situ, enabling integrative analysis of tissue organization across molecular layers. However, most existing graph-based integration methods rely on independently constructed modality-specific k-nearest-neighbor graphs. When auxiliary modalities are sparse or noisy, these graphs can become topologically discordant, propagate spurious edges, weaken cross-modal alignment, and reduce spatial domain resolution. RESULTS: We present Anchored RNA for Integrated Spatial Embedding (ARISE), an RNA expression anchored framework for spatial multi-omics integration. ARISE defines a shared-edge topology by intersecting RNA feature-similarity and spatial-proximity graphs, encodes auxiliary modalities on this common scaffold, and integrates them through inside-out hierarchical fusion. We further show theoretically that graph intersection minimizes false-positive edges within a broad class of k-of-r graph fusion rules, providing a principled basis for topology anchoring. Across various spatial multi-omics benchmarks spanning simulated and real datasets in bi-modal and tri-modal settings, ARISE improves spatial domain identification, cross-modal consistency, and preservation of tissue structure relative to existing methods. Furthermore, the learned representation supports biologically meaningful downstream analyses, including marker-based domain annotation, pathway enrichment, and cis-regulatory inference, indicating that ARISE yields a robust and interpretable framework for spatial multi-omics integration. AVAILABILITY AND IMPLEMENTATION: The source code is available at https://github.com/XiangxiangWang-code/ARISE. The archived version used in this study is available at https://doi.org/10.6084/m9.figshare.32686137.v2.

Multiomics↗
PubMed1995

Learning a visuomotor transformation in a local area of work space produces directional biases in other areas.

1. The dependence of directional biases in reaching movements on the initial position of the hand was studied in normal human subjects moving their unseen hand on a horizontal digitizing tablet to visual targets displayed on a vertical computer screen. 2. When initial hand positions were to the right of midline, movements were systematically biased clockwise. Biases were counterclockwise for starting points to the left. Biases were unaffected by the screen location of the starting and target positions. 3. Vision of the hand in relation to the target before movement, as well as practice with vision of the cursor during the movement, temporarily eliminated these biases. The spatial organization of the biases suggests that, without vision of the limb, the nervous system underestimates the distance of the hand from an axis or plane that includes its most common operating location. 4. To test the hypothesis that such an underestimate might represent an adaptation to a local area of work space or range effect, subjects were trained to reach accurately from right or left positions. After training, movements initiated from other locations, including ones that were previously error free, showed new biases that again represented underestimates of the distance of the initial hand position from the new trained location. 5. We conclude that hand path planning is dependent on learned representations of the location of the hand in the work space.

Adult↗
PubMed2026

Large language models in bioinformatics: a comprehensive survey.

The emergence of foundation models with trillion-level parameters has redefined the landscape of artificial intelligence. Various fields are developing their own large-scale models, which can solve many problems within the field and improve work efficiency. Biological large-scale models are a cross-disciplinary research field that combines mathematics, computer science, and biology, aiming to simulate and understand the structure, function, and dynamic changes of biological systems through the establishment of complex computational models. This field covers multiple levels such as biological pathways, population dynamics, protein folding, etc., providing us with tools for deep exploration of the mysteries of life and applications in medicine, ecology, and other fields. This article reviews the background and research status of biological large-scale models, and discusses future directions. Large language models (LLMs) and other large-scale foundation models have rapidly advanced in recent years, enabling powerful representation learning and generation across text, sequences, and multimodal data. In bioinformatics and biomedicine, these models are increasingly used to analyze genomic sequences, infer protein properties and structures, support drug discovery, and integrate heterogeneous biomedical evidence. This survey reviews the basic principles of LLMs and summarizes representative applications in (i) gene and genome sequence analysis, (ii) protein structure and function prediction, and (iii) drug design, including virtual screening and personalized medicine. We also discuss emerging multi-model modeling approaches, as well as key challenges such as data quality and privacy, interpretability, generalization to new organisms and tasks, and responsible deployment in health-related settings. Finally, we outline future directions for developing reliable, scalable, and explainable bioinformatics foundation models.

bioinformatics↗
PubMed2001

Are grammatical representations useful for learning from biological sequence data?--a case study.

This paper investigates whether Chomsky-like grammar representations are useful for learning cost-effective, comprehensible predictors of members of biological sequence families. The Inductive Logic Programming (ILP) Bayesian approach to learning from positive examples is used to generate a grammar for recognising a class of proteins known as human neuropeptide precursors (NPPs). Collectively, five of the co-authors of this paper, have extensive expertise on NPPs and general bioinformatics methods. Their motivation for generating a NPP grammar was that none of the existing bioinformatics methods could provide sufficient cost-savings during the search for new NPPs. Prior to this project experienced specialists at SmithKline Beecham had tried for many months to hand-code such a grammar but without success. Our best predictor makes the search for novel NPPs more than 100 times more efficient than randomly selecting proteins for synthesis and testing them for biological activity. As far as these authors are aware, this is both the first biological grammar learnt using ILP and the first real-world scientific application of the ILP Bayesian approach to learning from positive examples. A group of features is derived from this grammar. Other groups of features of NPPs are derived using other learning strategies. Amalgams of these groups are formed. A recognition model is generated for each amalgam using C4.5 and C4.5rules and its performance is measured using both predictive accuracy and a new cost function, Relative Advantage (RA). The highest RA was achieved by a model which includes grammar-derived features. This RA is significantly higher than the best RA achieved without the use of the grammar-derived features. Predictive accuracy is not a good measure of performance for this domain because it does not discriminate well between NPP recognition models: despite covering varying numbers of (the rare) positives, all the models are awarded a similar (high) score by predictive accuracy because they all exclude most of the abundant negatives.

Bayes Theorem↗
PubMed1996

Image representations for visual learning.

Computer vision researchers are developing new approaches to object recognition and detection that are based almost directly on images and avoid the use of intermediate three-dimensional models. Many of these techniques depend on a representation of images that induce a linear vector space structure and in principle requires dense feature correspondence. This image representation allows the use of learning techniques for the analysis of images (for computer vision) as well as for the synthesis of images (for computer graphics).

Artificial Intelligence↗
PubMed2001

Linear constraints on weight representation for generalized learning of multilayer networks.

In this article, we present a technique to improve the generalization ability of multilayer neural networks. The proposed method introduces linear constraints on weight representation based on the invariance natures of training targets. We propose a learning method that introduces effective linear constraints into an error function as a penalty term. Furthermore, introduction of such constraints leads to reduction of the VC dimension of neural networks. We show bounds on the VC dimension of the neural networks with such constraints. Finally, we demonstrate the effectiveness of the proposed method by some experiments.

Algorithms↗
PubMed1994

[Experimental test of a model of memory-representation-generation in learning and recognition].

A highly structured set of stimuli was used in this study. Each stimulus had four binary attributes, whose values were determined so that any two stimuli could be transformed into each other by changing values of one or more attributes. In one experiment, 93 undergraduates rated similarity of paired stimuli. In another experiment, the same subjects learned three stimuli which were presented one after another for 10 seconds each. Later, in the recognition task, they made "old" or "new" judgment and rated the confidence of their judgment for each of the test stimuli. Two groups of subjects served the two experiments in different order. The results showed that (1) the rated similarity between the paired stimuli is a monotonically decreasing function of the number of transformations needed to get the pair equal, (2) the recognition confidence for new stimuli is significantly higher for stimuli generated by relevant transformations from the learned stimuli than for stimuli not so generated. The results support a model of memory-representation-generation (Suto, 1987, 1988), but not "prototype plus transformation model" nor "context model".

Adult↗
PubMed2002

Associative learning and elemental representation: II. Generalization and discrimination.

This paper follows on from an earlier companion paper (McLaren & Mackintosh, 2000), in which we further developed the elemental associative theory put forward in McLaren, Kaye, and Mackintosh (1989). Here, we begin by explicating the idea that stimuli can be represented as patterns of activation distributed across a set of units and that different stimuli activate partially overlapping sets (the degree of overlap being proportional to the similarity of the stimuli). A consequence of this view is that the overall level of activity of some of the units representing a stimulus may be dependent on the nature of the other stimuli present at the same time. This allows an elemental analysis in which provision for the representation of configurations of stimuli is made. A selective review of studies of generalization and discrimination learning, including peak shift, transfer along a continuum, configural discrimination, and summation, suggests that the principles embodied in this class of theory deserve careful consideration and will form part of any successful model of associative learning in humans or animals. There are some phenomena that require an elemental/associative explanation.

Association Learning↗
PubMed1992

Combining exemplar-based category representations and connectionist learning rules.

Adaptive network and exemplar-similarity models were compared on their ability to predict category learning and transfer data. An exemplar-based network (Kruschke, 1990a, 1990b, 1992) that combines key aspects of both modeling approaches was also tested. The exemplar-based network incorporates an exemplar-based category representation in which exemplars become associated to categories through the same error-driven, interactive learning rules that are assumed in standard adaptive networks. Experiment 1, which partially replicated and extended the probabilistic classification learning paradigm of Gluck and Bower (1988a), demonstrated the importance of an error-driven learning rule. Experiment 2, which extended the classification learning paradigm of Medin and Schaffer (1978) that discriminated between exemplar and prototype models, demonstrated the importance of an exemplar-based category representation. Only the exemplar-based network accounted for all the major qualitative phenomena; it also achieved good quantitative predictions of the learning and transfer data in both experiments.

Concept Formation↗
PubMed1990

Hippocampal representation in place learning.

The generality of the place-learning impairment associated with hippocampal system damage was challenged using methods of training that permitted subjects to form an individual association between the place of escape and a particular navigational route in an open-field water maze. Both normal rats and rats with fornix lesions (FX rats) acquired this task rapidly, although FX rats were slightly slower in achieving minimum escape latencies. In postcriterion testing, FX rats occasionally made near misses but, more often, their escape performance was indistinguishable from that of intact rats. Results from a variety of probe tests indicated that FX rats, like normal rats, had based their performance on a representation of multiple distal cues but their representation, unlike that of normal rats, was inflexible in that it could not be used to guide performance when the cues or starting position were altered. These results parallel those from other studies of hippocampal function in animals and humans: The learning deficit consequent to hippocampal system damage (1) is not specific to a particular category of learning materials, but is dependent on the representational demands of the task; (2) is observed when task demands encourage a representation based on relations among multiple cues, but not when the task encourages adaptation to an individual (or compound) stimulus; (3) spares acquisition of fundamental procedures needed to perform the task; and (4) impairs the flexible use of learned information in tests other than repetition of the learning experience.

Animals↗
PubMed2001

A variational method for learning sparse and overcomplete representations.

An expectation-maximization algorithm for learning sparse and overcomplete data representations is presented. The proposed algorithm exploits a variational approximation to a range of heavy-tailed distributions whose limit is the Laplacian. A rigorous lower bound on the sparse prior distribution is derived, which enables the analytic marginalization of a lower bound on the data likelihood. This lower bound enables the development of an expectation-maximization algorithm for learning the overcomplete basis vectors and inferring the most probable basis coefficients.

Journal Article↗
PubMed2003

Dictionary learning algorithms for sparse representation.

Algorithms for data-driven learning of domain-specific overcomplete dictionaries are developed to obtain maximum likelihood and maximum a posteriori dictionary estimates based on the use of Bayesian models with concave/Schur-concave (CSC) negative log priors. Such priors are appropriate for obtaining sparse representations of environmental signals within an appropriately chosen (environmentally matched) dictionary. The elements of the dictionary can be interpreted as concepts, features, or words capable of succinct expression of events encountered in the environment (the source of the measured signals). This is a generalization of vector quantization in that one is interested in a description involving a few dictionary entries (the proverbial "25 words or less"), but not necessarily as succinct as one entry. To learn an environmentally adapted dictionary capable of concise expression of signals generated by the environment, we develop algorithms that iterate between a representative set of sparse representations found by variants of FOCUSS and an update of the dictionary using these sparse representations. Experiments were performed using synthetic data and natural images. For complete dictionaries, we demonstrate that our algorithms have improved performance over other independent component analysis (ICA) methods, measured in terms of signal-to-noise ratios of separated sources. In the overcomplete case, we show that the true underlying dictionary and sparse sources can be accurately recovered. In tests with natural images, learned overcomplete dictionaries are shown to have higher coding efficiency than complete dictionaries; that is, images encoded with an overcomplete dictionary have both higher compression (fewer bits per pixel) and higher accuracy (lower mean square error).

Algorithms↗