Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Generative models”

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
PubMed1977

A second order model of the optic generator potential and its relation to Stevens' power function.

The component PIII of the electroretinogram representing the optic generator potential was recorded after stimulation with short stimuli with different light intensity. It is shown that the impulse response function of a linear second order model with intensity-dependent coefficients can be well fit to the recordings. Two of its parameters, after logarithmic transformations, are linearly dependent on the luminance while the third parameter varies only within a small interval. It is therefore possible to describe the relation between PIII and luminance in a linearized second order model. Furthermore, both the type of the function relating the gain of the model to the luminance and its exponent are nearly identical with the psychophysic function relating luminance to subjective brightness. Further physiological implications are also considered.

Animals↗
PubMedDate not supplied

[Mechanism of early receptor potential generation and an electrical model of retinal rods in rats].

A passive electric model of the retinal rod can reproduce the changes of the ELP-waveform at temperatures within the range 10 degrees--45 degrees C. R2-component of the ERP is generated by the process with the time constant ca. 500 microseconds (37 degrees C) and activation energy 31 kcal/mol characteristic of metarhodopsin I--metarhodopsin II transition. R1 originates not later than at prelumirhodopsin--lumirhodopsin conversion. Charge displacements are 0.006 (R1) and 0.04 (R2) electronic charges per rhodopsin bleached. The model is used for determining membrane capacitance and resistance of the rat retinal rod.

Animals↗
PubMed2026

A Patient-Derived Xenograft Repository Capturing Clinical and Molecular Heterogeneity of Large B-cell Lymphoma.

UNLABELLED: Large B-cell lymphomas (LBCL) are a clinically and molecularly diverse group of malignancies with a rapidly evolving therapeutic landscape that has introduced new areas of clinical need, such as post-CD19 chimeric antigen receptor T (CART19) progression. Patient-derived xenograft (PDX) models are an important tool for mechanistic studies and preclinical evaluation of new therapies and can be generated from a variety of clinical contexts that capture tumor-intrinsic resistance mechanisms. We therefore undertook a comprehensive effort to generate PDX models that encompass the molecular landscape of LBCLs and include important clinical scenarios for new drug development. Here, we describe the first 48 models within this publicly available repository, capturing the transcriptional and genetic subsets of LBCL. These models also include 23 generated from post-CART19 progression patient biopsies, which reproduce patterns of progression driven by CD19 mutation or expression loss, as well as tumor cell-intrinsic CART19 resistance that we validated in vivo. SIGNIFICANCE: Here, we describe X-LYMPH (Xenografts of Lymphoma), a publicly available and molecularly annotated PDX repository that captures the heterogeneity of LBCL. X-LYMPH includes models of CAR T-cell resistance, providing a shared foundation for mechanistic research and therapeutic development for lymphomas. See related commentary by Evgin and Steidl, p. 655.

Humans↗
PubMed1979

Synthesis and degradation of xanthine dehydrogenase in chick liver. In vivo and in vitro studies.

The present study describes the (xanthine:NAD+ oxidoreductase, EC 1.2.1.37) synthesis and degradation of chick liver xanthine dehydrogenase in vivo and in organ cultures. The results indicate that control of xanthine dehydrogenase activity is mediated by changes in the rate of enzyme synthesis, but that degradation rates are unaffected. The results also suggest that xanthine dehydrogenase synthesis occurs through a previously unreported intermediate. Detected in cultures of liver tissue, this intermediate apparently is not converted into an active enzyme. A model of synthesis and degradation for xanthine dehydrogenase proposes that the synthesis of the enzyme is proportional to messenger RNA and includes an inactive enzyme precursor and a second inactive intermediate prior to degradation. Integrated mathematical solutions describing the concentration of intermediates as a function of time can be found explicitly for simple models. The appendix to this paper extrapolates solutions for one-, two- and three-step models to generate a mathematical solution for an 'n'-step model containing 'n' intermediates. The rate constants in the solutions can be found experimentally.

Animals↗
PubMed1975

Thermal fields and heat generation effects in tissue, awake and under halothane anesthesia.

In order to further document the heat generation terms used in models predicting heat transmission in tissue, a series of temperature fields were generated and measured in cat brain with an implantable cylindrical source and thermocouple apparatus. The experimental technique allowed measurements to be made under fully awake in vivo conditions as well as for 3 to 45% levels of halothane anesthesia. Fields resulting from probe forcing function temperatures of 4 degrees C, 18 degrees C, and 28 degrees C were warmer in the anesthetized tissue indicating an increase in heat generation resulting from blood flow and metabolic heat. Values of the heat generation were obtained by applying the Bio-Heat Equation directly to the experimental data. A simple blood flow heat generation model gave a reasonable prediction of the heat generation in a temperature range lower than 3 to 4 degrees C below the inlet blood temperature, and with a Q10 equals 3 variation of blood flow rate with temperature. Results also indicate that the heat generation effect under halothane anesthesia conditions are such that the use of the apparent property concept in thermal modeling may not be valid.

Anesthesia, Inhalation↗
PubMed2025

C6ORF120 regulates hepatic lipid metabolism through PPAR signaling pathway in metabolic dysfunction-associated steatotic liver disease.

Background Emerging evidence indicates that C6ORF120 is highly expressed in the liver and may modulate immune responses in various hepatic disorders. However, its role in hepatic lipid metabolism and metabolic dysfunction-associated steatotic liver disease (MASLD) is unexplored. This study aimed to elucidate the effects and potential mechanisms of C6ORF120 on hepatic lipogenesis. Methods C6ORF120 expression in MASLD was assessed using patient serum and the Gene Expression Omnibus (GEO) database. A high-fat diet-induced MASLD model was established in C6orf120-KO rats. Fatty acid-induced lipid accumulation models were generated in primary hepatocytes, HepG2 and Huh7 cells. These models were employed to investigate the effects of C6ORF120 on hepatic lipogenesis and MASLD progression. Results C6ORF120 expression was significantly upregulated in MASLD patients and obese rat models. Genetic deletion of C6ORF120 markedly alleviated high-fat diet-induced steatosis in the liver of rats. In vitro, C6orf120 gene deficiency attenuated lipid accumulation and suppressed key lipogenic genes (such as fatty acid synthase (Fasn), phospho-acetyl coenzyme carboxylase (p-ACC), sterol regulatory element binding protein-1c (Srebp1c)) in primary hepatocytes and HepG2 cells. Conversely, C6ORF120 overexpression increased lipid accumulation in HepG2 cells. RNA sequencing analysis showed that lipid metabolism pathway and peroxisome proliferators activated receptor (PPAR) signaling pathway were significantly altered in the liver of C6orf120-KO rats. We demonstrated that C6ORF120 may regulate lipid metabolism through the hepatic PPARα, which is involved in fatty acid production and lipid oxidation. Further, we found that serum C6ORF120 expression was correlated with clinical indicators in patients with MASLD. Conclusion This study preliminarily revealed a novel function for C6ORF120 in hepatic lipid metabolism via affecting the PPAR pathway. The result identifies C6ORF120 as a novel regulator of hepatic lipid metabolism through PPARα-dependent mechanisms, offering potential therapeutic targets for MASLD.

Lipid Metabolism↗
PubMedDate not supplied

[Effect of the curvature of the dorsal hippocampus on the shape of the electric field which it generates].

A mathematical model is presented of spatial configuration of the field generated by hippocampus surface. Calculations have shown that the potential induced on the cortex surface is almost 5-fold smaller than that under the hippocampus surface. Theoretical results well agree with the experimental data and allow a conclusion to be inferred that passive penetration of fields generated by subcortical structures does not exceed 20% of the total level of spontaneous EEG.

Electroencephalography↗
PubMed1977

[Modeling neuropathologic syndromes by creating generators of pathologically enhanced excitation in the hypothalamus of rabbits].

In the experiments on free behavior rabbits, tetanus toxin was injected into "pacemaker" motivational emotiogenic regions of the hypothalamus to form generators of pathologically enhanced excitation; this produced stable, long-term disorders in motivational-emotional behavior. The changes were manifested by intensification of the feeding behavior activity, including increase of the "secondary motivational reactions", intensification of the motor activity, excessive number of automatic masticatory movements, appearance of aggression, fear reaction and corresponding vegetative changes. The character of these reactions depended on the site of the toxin administration and on its dose. Formation of long-term generators of the pathologically enhanced excitation in the "pacemaker" motivational-emotiogenic centers of the hypothalamus by tetanus toxin can be used the modelling of psychopathological states in animals. The data obtained on the new model have confirmed the theory of generative mechanisms of neuropathological syndromes characterized by hyperactivity of the systems.

Aggression↗
PubMed2026

ChromBERT-tools: a versatile toolkit for context-specific regulatory representations of transcription regulators across different cell types.

SUMMARY: Representations that encode the genome-wide regulatory behavior of transcription regulators provide a foundation for flexible transcription modeling and in silico regulatory analysis. Existing regulator representations are commonly derived from gene co-expression, motif annotations, or static protein features, which capture useful but limited aspects of regulator identity but do not directly model how regulators participate in region-specific regulatory programs across the genome. ChromBERT addresses this gap by learning context-aware regulatory representations from large-scale ChIP-seq data. However, routine bioinformatics applications require lightweight, accessible, and modular tools for generating, adapting, and interpreting these representations in user-defined biological contexts. Here, we present ChromBERT-tools, a user-oriented toolkit built upon ChromBERT that converts its regulatory representation framework into practical workflows for customizable analysis across cellular contexts. ChromBERT-tools provides command-line interfaces and Python APIs organized into three functional layers: representation generation, predictive modeling, and regulatory interpretation. The representation generation layer produces representations of genomic regions and transcription regulators. The predictive modeling layer fine-tunes ChromBERT for genome-wide regulatory activity prediction through classification or regression tasks, with optimized implementation to reduce running time and computational resource requirements. The regulatory interpretation layer supports inference of the context-specific roles of cis-regulatory elements and transcription regulators. These modules can be used independently or integrated into end-to-end workflows, enabling flexible analyses across diverse datasets. ChromBERT-tools lowers the barrier to applying context-specific regulatory representations in routine genomic analyses. AVAILABILITY AND IMPLEMENTATION: ChromBERT-tools is freely available at https://github.com/TongjiZhanglab/ChromBERT-tools, with documentation at https://chrombert-tools.readthedocs.io/en/latest/. A frozen archival snapshot is available on Zenodo under DOI: 10.5281/zenodo.20094206.

Software↗
PubMed2023

De novo variants in MRTFB have gain-of-function activity in Drosophila and are associated with a novel neurodevelopmental phenotype with dysmorphic features.

PURPOSE: Myocardin-related transcription factor B (MRTFB) is an important transcriptional regulator, which promotes the activity of an estimated 300 genes but is not known to underlie a Mendelian disorder. METHODS: Probands were identified through the efforts of the Undiagnosed Disease Network. Because the MRTFB protein is highly conserved between vertebrate and invertebrate model organisms, we generated a humanized Drosophila model expressing the human MRTFB protein in the same spatial and temporal pattern as the fly gene. Actin binding assays were used to validate the effect of the variants on MRTFB. RESULTS: Here, we report 2 pediatric probands with de novo variants in MRTFB (p.R104G and p.A91P) and mild dysmorphic features, intellectual disability, global developmental delays, speech apraxia, and impulse control issues. Expression of the variants within wing tissues of a fruit fly model resulted in changes in wing morphology. The MRTFBR104G and MRTFBA91P variants also display a decreased level of actin binding within critical RPEL domains, resulting in increased transcriptional activity and changes in the organization of the actin cytoskeleton. CONCLUSION: The MRTFBR104G and MRTFBA91P variants affect the regulation of the protein and underlie a novel neurodevelopmental disorder. Overall, our data suggest that these variants act as a gain of function.

Animals↗
PubMed2026

Bridging ancestry gaps in genomic risk prediction with tabular foundation models.

MOTIVATION: Models deployed for genomic prediction of diseases perform unevenly across populations, limiting clinical utility. Two factors drive this limitation: large imbalances in sample availability across ancestry groups and non-stationarity of genotype-phenotype effect sizes across the ancestry continuum. While tabular foundation models with in-context learning (ICL) have shown strong sample efficiency in other domains, their effectiveness for genotype-to-phenotype prediction and their robustness to ancestry-driven effect heterogeneity remain unclear. RESULTS: Using large, ancestrally diverse biobank data, we show that ICL-capable tabular foundation models reduce performance degradation in under-sampled ancestry groups compared to conventional supervised approaches. However, we find that prevailing models trained on existing synthetic tabular tasks fail when allele effect sizes vary across ancestry space. Treating genetic ancestry as a continuous variable, we introduce an instruction-tuning framework that exposes models to synthetic tasks with ancestry-dependent non-stationary effects. Instruction-tuned models achieve improved and more stable predictive performance across the genetic ancestry continuum, including for individuals distant from in-context exemplars in ancestry space. AVAILABILITY AND IMPLEMENTATION: All code for instruction-tuning models, synthetic task generation, data wrangling, and model evaluation, is publicly available at https://github.com/ai4pm/Bridging-Ancestry-Gaps-in-Genomic-Risk-Prediction-with-Tabular-Foundation-Models. The final instruction-tuned model (ICL-NS-G2P-proto) is also released in this repository. Detailed documentation is provided, including environment setup instructions and guidelines for running various parts. The instruction-tuning task datasets are available at https://zenodo.org/records/18309187.

Humans↗
PubMed2026

Bridging Ancestry Gaps in Genomic Risk Prediction with Tabular Foundation Models.

MOTIVATION: Models deployed for genomic prediction of diseases perform unevenly across populations, limiting clinical utility. Two factors drive this limitation: large imbalances in sample availability across ancestry groups and non-stationarity of genotype-phenotype effect sizes across the ancestry continuum. While tabular foundation models with in-context learning (ICL) have shown strong sample efficiency in other domains, their effectiveness for genotype-to-phenotype prediction and their robustness to ancestry-driven effect heterogeneity remain unclear. RESULTS: Using large, ancestrally diverse biobank data, we show that ICL-capable tabular foundation models reduce performance degradation in under-sampled ancestry groups compared to conventional supervised approaches. However, we find that prevailing models trained on existing synthetic tabular tasks fail when allele effect sizes vary across ancestry space. Treating genetic ancestry as a continuous variable, we introduce an instruction-tuning framework that exposes models to synthetic tasks with ancestry-dependent non-stationary effects. Instruction-tuned models achieve improved and more stable predictive performance across the genetic ancestry continuum, including for individuals distant from in-context exemplars in ancestry space. AVAILABILITY AND IMPLEMENTATION: All code for instruction-tuning models, synthetic task generation, data wrangling, and model evaluation, is publicly available at https://github.com/ai4pm/Bridging-Ancestry-Gaps-in-Genomic-Risk-Prediction-with-Tabular-Foundation-Models. The final instruction-tuned model (ICL-NS-G2P-proto) is also released in this repository. Detailed documentation is provided, including environment setup instructions and guidelines for running various parts. The instruction-tuning task datasets are available at https://zenodo.org/records/18309187.

Ancestry Continuum↗
PubMed1978

Tuberculosis: generation effects and chemotherapy.

Mortality from pulmonary tuberculosis in the United States was analyzed by cohort. The introduction of effective chemotherapy necessarily renders models based on generation differences alone inappropriate in this disease. However, such models continue to be used. The data show major departures from the prediction of a generation based model during the 1950's. Projection of the 1941 rates and cohort slopes to 1970 using a generation model predicts 358,000 more deaths than were actually certified. The departure was smaller for blacks than for whites, and differences in delivery of treatment probably account for this. Projections of mortality need to be made for the planning of control measures. However, such projections must be done with the understanding that inter-cohort differences can be altered during adult life.

Adolescent↗
PubMed2026

A Clinically Integrated Pediatric Patient-Derived Xenograft Program Enables Evaluation of Cohort and Patient-Specific Biology and Therapeutic Strategies.

UNLABELLED: Preclinical translational research has increasingly utilized patient-derived xenograft (PDX) models for mechanistic and experimental therapeutic studies, yet most existing models have been developed from adult cancer types. We describe the establishment of a PDX program to expand the availability of pediatric-specific PDXs for preclinical research and enable studies of pediatric cancer histologies, including ultrarare diseases. Processes for PDX generation were integrated into established clinical workflows to facilitate universal model generation. Methodologies for tissue procurement, processing, and cryopreservation were optimized to enable intra- and interinstitutional PDX model generation. Over a 6-year span, 388 PDX tumor models representing more than 40 diagnoses were generated, including ultrarare tumors and longitudinal models established from pretherapy, posttherapy, and relapse tumors from the same patient. Genomic characterization of these PDXs demonstrates excellent concordance and recapitulation of molecular alterations of the source tumor. Successful PDX generation was enhanced from relapsed samples, was higher in sarcomas compared with other solid tumor types, and was a negative prognosticator for clinical outcome. With a broad portfolio of molecularly annotated models, we demonstrate utility for validating cross-histology biomarker-driven therapeutic strategies by demonstrating antitumor activity of an MAT2A inhibitor in MTAP-deficient PDXs. Universal model creation also allows for experimental validation of therapeutic hypotheses on a patient-specific basis, as we describe the characterization of a novel RAF1 fusion (EPB41L2::RAF1) in an osteosarcoma PDX. Development of a diverse collection of pediatric PDX models enables hypothesis-driven and cross-histology studies that expand our understanding of cancer biology and aid ongoing drug prioritization efforts in rare tumors. SIGNIFICANCE: A clinically integrated, genomically annotated pediatric PDX portfolio supported by systematic benchmarking of model generation facilitates exploratory biomarker-driven and patient-specific translational studies.

Humans↗
PubMed1978

Systems-matching by degeneration. II. Interpretation of the generation and degeneration of retinal ganglion cells in the chicken by a mathematical model.

Quantitative data on generation and degeneration of retinal ganglion cells during development (Rager and Rager, 1978) are interpreted in terms of a mathematical model which consists of a system of differential equations. By these equations we attempt to describe the formation of retinal ganglion cells and their termination domains in the tectum. Since ganglion cells seem not to degenerate before their axons have arrived at their termination site and start branching, from the arrival time on they may become competent either to continue to mature or to die. Therefore, to find the actual number of competent cells the extension of the fiber pathway between the retina and the optic tectum had also to be measured and computed. The differential equations are united by the principle that at any given time cells in excess of the number of termination domains have to die. By this model the mathematical function was determined. Several parameter values of this function were optimized with the Gauss-Newton method by which the curve was fitted to the measured values. The high correlation obtained by this method allows to conclude that, to a first approximation, the model may be satisfactory. The evidence of competition for termination sites and of systems-matching by cell death is discussed.

Age Factors↗
PubMed1975

Programmatic computer simulation model for medical school planning.

A comprehensive programmatic computer simulation planning model for a school of medicine was generated by integrating several separate simulation models with on-site cost study information. An elementary validation of the model was achieved. The model generated program costs in terms of both faculty hours and dollars. Results indicated that the size of the medical class could be increased from 75 to 100 students within the present resource limitations by transferring faculty time to education from other programs. The maximum class size was limited by the availability of clinical material. The basic science departments could handle this class size easily without significant reduction in other programs, but the clinical departments could not do so unless inpatient levels increased significantly.

Computers↗