PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “reinforcement learning”

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 19 recordsLinked to original sources
PubMed1979

Aspects of the reinforcer learned in second-order Pavlovian conditioning.

Four experiments used an autoshaping procedure in pigeons to explore learning about the reinforcer in a second-order conditioning paradigm. Experiment 1 conditioned two visual second-order stimuli (S2), using as reinforcers two visual first-order stimuli (S1), each of which had previously been paired with food. Animals for which the S2 stimuli were each consistently paired with one particular S1 developed second-order responding more rapidly than did animals for which the identity of S1 varied from trial to trial. Moreover, following consistent pairings, extinction of an S1 had a depressive effect upon second-order responding which was peculiar to the S2 with which it had been paired. Both results suggest that in this preparation the organism identifies a particular S1 as the reinforcer for each S2. The remaining experiments examined the details of that identification. A compound S1, itself composed of two separable elements, was used to reinforce an S2. Subsequent extinction of either element of S1 led to a depression in the responding to S2, which indicates that both elements were involved in the second-order conditioning. Moreover, the use of several complex discriminations, which produced different behavior to S1 and to its elements, suggested that the organism had associated the S2 with the compound S1 rather than with its separate elements. However, even complete extinction of the response to S1 left some residual behavior to S2, which indicates that a portion of the second-order conditioning is independent of the current state of the reinforcer. These results demonstrate that in some situations the organism associates a conditioned stimulus with a rich representation of the reinforcer.

Animals↗
PubMed2026

CRISPGen: A deep generative framework for multi-objective CRISPR/Cas9 guide RNA design via Conditional Latent Diffusion and Dual-Critic Reinforcement Learning.

MOTIVATION: The CRISPR-Cas9 system offers transformative potential for precision genome editing, yet its clinical translation remains constrained by the risk of unintended off-target double-strand breaks. While current discriminative models excel at evaluating pre-specified candidate guides, resolving the fundamental antagonism between on-target cleavage efficiency and off-target specificity within a fixed sequence search space remains a major challenge. RESULTS: We present CRISPGen, a unified deep generative framework that reframes sgRNA design as a multi-objective constrained sequence synthesis problem. It integrates (i) DNABERT-2 genomic-language embeddings, (ii) a conditional latent diffusion generator conditioned on a user-specified on-target efficiency target, and (iii) a dual-critic reinforcement-learning (RL) stage that couples a frozen on-target efficiency critic with a cross-attention off-target discriminator (validation Pearson R=0.8157) trained on a unified corpus of experimental off-target events from six detection platforms. Across 1000 generated sgRNAs, CRISPGen reduces the mean off-target discriminator score by 99.7% relative to the pre-RL baseline and, under an exhaustive whole-genome screen of all 302,631,056 NGG PAM sites in GRCh38, yields zero perfect-match and only 55 one-mismatch genomic hits. We further show, transparently, that the internal on-target critic saturates under RL optimization - an instance of Goodhart's Law - and therefore assess on-target viability using an independent external CRISPRon screen (mean 47.10/100). Repeating the RL fine-tuning stage under three random seeds (with the diffusion generator, DNABERT-2 embeddings, and off-target discriminator held fixed) yields a stable operating point across seeds. Full diversity, per-mismatch, and reproducibility statistics are reported in the Results. AVAILABILITY: Source code is available at https://github.com/malekpouri/CRISPGen; the pre-trained checkpoints and the 3,000,000-sequence library are hosted on Hugging Face (https://huggingface.co/malekpouri/CRISPGen-Checkpoints) and archived on Zenodo under DOI 10.5281/zenodo.21428641.

CRISPR-Cas9↗
PubMed2025

GiantHunter: accurate detection of giant virus in metagenomic data using reinforcement-learning and Monte Carlo tree search.

MOTIVATION: Nucleocytoplasmic large DNA viruses (NCLDVs) are notable for their large genomes and extensive gene repertoires, which contribute to their widespread environmental presence and critical roles in processes such as host metabolic reprogramming and nutrient cycling. Metagenomic sequencing has emerged as a powerful tool for uncovering novel NCLDVs in environmental samples. However, identifying NCLDV sequences in metagenomic data remains challenging due to their high genomic diversity, limited reference genomes, and shared regions with other microbes. Existing alignment-based and machine learning methods struggle with achieving optimal trade-offs between sensitivity and precision. RESULTS: In this work, we present GiantHunter, a reinforcement learning-based tool for identifying NCLDVs from metagenomic data. By employing a Monte Carlo tree search strategy, GiantHunter dynamically selects representative non-NCLDV sequences as the negative training data, enabling the model to establish a robust decision boundary. Benchmarking on rigorously designed experiments shows that GiantHunter achieves high precision while maintaining competitive sensitivity, improving the F1-score by 10% and reducing computational cost by 90% compared to the second-best method. To demonstrate its real-world utility, we applied GiantHunter to 60 metagenomic datasets collected from six cities along the Yangtze River, located both upstream and downstream of the Three Gorges Dam. The results reveal significant differences in NCLDV diversity correlated with proximity to the dam, likely influenced by reduced flow velocity caused by the dam. These findings highlight GiantHunter's potential to advance our understanding of NCLDVs and their ecological roles in diverse environments. AVAILABILITY AND IMPLEMENTATION: The source code of GiantHunter is available via: https://github.com/FuchuanQu/GiantHunter.

Metagenomics↗
PubMed2026

Reinforcement learning-based dynamic ensemble for missense variant effect prediction and tiered prioritization of VUS.

BACKGROUND: Accurate classification of missense variants remains a challenging task despite major advances in genomics. Numerous computational models have been developed to assist in variant classification, but often require repeated integration and benchmarking efforts. Ensemble methods have been proposed to overcome the limitations of single predictors, but mostly rely on fixed, predefined weights that constrain their ability to capture interactions among predictive signals. METHODS: We present GenixRL, a dynamic ensemble framework that reformulates model fusion as a reinforcement learning optimization problem. GenixRL uses a Q-learning agent to learn a policy that dynamically weights the probabilistic outputs of complementary predictors, including BayesDel (addAF and noAF), ClinPred, and MetaRNN. Replacing static weighting with policy learning allows GenixRL to adaptively identify optimal weightings and substantially improve classification accuracy. RESULTS: In benchmark evaluation against 25 state-of-the-art predictors, GenixRL achieved an AUROC of 0.9644 on an independent ClinVar dataset. On saturation genome editing assays for BRCA1 and BRCA2, GenixRL achieved the best performance and ranked highest on 14 of 17 clinically significant genes in a zero-shot evaluation. Applied to uncertain and conflicting ClinVar variants, GenixRL enabled tiered, evidence-based prioritization of hundreds of thousands of variants as likely pathogenic or pathogenic with high confidence, supported by orthogonal population evidence from gnomAD. CONCLUSION: GenixRL advances pathogenicity prediction for missense variants and provides an adaptive ensemble that sorts variants of uncertain significance into tiered candidates for expert curation and functional validation.

Mutation, Missense↗
PubMed2026

A reinforcement learning-enhanced fuzzy multi-objective equilibrium optimization framework for multiple sequence alignment.

Multiple sequence alignment (MSA) is a fundamental task in bioinformatics, underpinning comparative genomics, structural analysis, and evolutionary inference. However, MSA remains a challenging multi-objective optimization problem due to the need to simultaneously maximize alignment accuracy, preserve conserved regions, and control gap proliferation, particularly in large and heterogeneous sequence collections. In this work, we propose MOFSACEO-MSA, a novel hybrid optimization framework for multiple sequence alignment that integrates a fuzzy multi-objective evaluation scheme with the Equilibrium Optimizer (EO) and a Soft Actor-Critic (SAC)-based adaptive control mechanism. The proposed framework formulates MSA as a dynamic multi-objective optimization problem, in which alignment quality is assessed using complementary residue-level and column-level criteria, including Sum-of-Pairs score, column conservation, entropy, and gap statistics. Fuzzy membership functions are employed to harmonize competing objectives into a unified optimization landscape, while EO provides robust global exploration. To further enhance adaptability, SAC dynamically regulates key EO parameters during the search process, enabling an effective balance between exploration and exploitation across datasets of varying size and heterogeneity. Extensive experiments werew conducted on diverse biological sequence datasets, with a primary focus on RNA benchmarks, including structured families from Rfam, large-scale repositories from RNAcentral and GenBank, and organism-specific tRNA datasets from GtRNAdb. Comparative evaluations against classical alignment tools (ClustalW, MAFFT, MUSCLE, PRANK, KAlign, and T-Coffee), metaheuristic methods (SAGA, Sequoya and EAFSA), and a reinforcement learning-based approach (RLALIGN) demonstrate that MOFSACEO-MSA consistently achieves competitive or superior Sum-of-Pairs scores while significantly reducing gap proportions and maintaining compact alignment lengths. Notably, the proposed framework exhibits improved robustness on large and highly heterogeneous datasets, where existing methods often suffer from excessive gap insertion or unstable convergence. Overall, MOFSACEO-MSA provides a flexible and extensible optimization paradigm that effectively bridges evolutionary search and reinforcement learning for high-quality multiple sequence alignment, with demonstrated effectiveness on challenging RNA alignment tasks.

Sequence Alignment↗
PubMed1978

Evolution of the Gynecology Teaching Associate: an education specialist.

The traditional pelvic examination instruction methods were reviewed and found to be deficient: the student learning experience was compromised by the triangular setting of patient, student, and instructor for early pelvic examination instruction. Over the past decade, a new education specialist, the Gynecology Teaching Associate (GTA), has evolved to help improve the initial gynecology teaching experience. The evolution of the GTA is described. The qualities she brings to the instructional system include sensitivity as a woman, educational skill in pelvic examination instruction, knowledge of female pelvic anatomy and physiology, and, most important, sophisticated interpersonal skills to help medical students learnin in a nonthreatening environment. Reinforcement learning theory is the foundation of this educational system. Student acceptance of this system is documented.

Curriculum↗
PubMed2026

Mechanistic insights into Claudin-14 dysfunction implicated in veins of Galen malformation.

Claudin-14 (CLDN14) is a key component of tight junctions (TJs) critical for maintaining paracellular barrier function. Variants of CLDN14 have been linked to Vein of Galen malformations (VOGMs), a rare cerebrovascular disorder; however, the molecular mechanisms underlying their pathogenicity remain unknown. Here, we investigate the mechanistic effects of two VOGM-associated mutations, A113P and V143M, using reinforcement-learning driven enhanced sampling molecular dynamics simulations combined with DiffNets-based deep learning and independent trajectory-wide structural analyses. Our analysis reveals that A113P induces broader structural disruption of CLDN14, perturbing paracellular sealing, pore symmetry, and inter-protomer communication, whereas V143M induces structural rearrangements centred around TM3 and the TM3-ECL2 region. In both cases, mutation-specific alterations are observed in structural stability and interfacial organization across oligomeric assemblies. Notably, these effects are qualitatively consistent across different modelled architectures, despite variability in local responses. In the absence of experimentally resolved structures, the structural perturbations reported here provide a mechanistic understanding of how VOGM-associated variants may influence CLDN14 structure and dynamics.

Aneurysm↗
PubMed2025

Asymmetric cortical projections to striatal direct and indirect pathways distinctly control actions.

The striatal direct and indirect pathways constitute the core for basal ganglia function in action control. Although both striatal D1- and D2-spiny projection neurons (SPNs) receive excitatory inputs from the cerebral cortex, whether or not they share inputs from the same cortical neurons, and how pathway-specific corticostriatal projections control behavior remain largely unknown. Here using a G-deleted rabies system in mice, we found that more than two-thirds of excitatory inputs to D2-SPNs also target D1-SPNs, while only one-third do so vice versa. Optogenetic stimulation of striatal D1- vs. D2-SPN-projecting cortical neurons differently regulate locomotion, reinforcement learning and sequence behavior, implying the functional dichotomy of pathway-specific corticostriatal subcircuits. These results reveal the partially segregated yet asymmetrically overlapping cortical projections on striatal D1- vs. D2-SPNs, and that the pathway-specific corticostriatal subcircuits distinctly control behavior. It has important implications in a wide range of neurological and psychiatric diseases affecting cortico-basal ganglia circuitry.

Journal Article↗
PubMed1978

[Effects of bilateral amygdalectomy on levers pressing for food reward in the rabbit (author's transl)].

Effects for continuous reinforcement (CRF) learning and differential reinforcement of low rate (DRL) learning for food in the rabbit by bilateral amygdalectomy were examined with the Skinner box. There was no changes in general behaviour and no impediment of CRF learning. But the completion of DRL learning was delayed by amygdalectomy. In the stage of completion of DRL learning there was no difference of total number of the reinforcement and the lever pressing between the control group and amygdalectmized group. But there was the impediment of DRL learning by the analysis of frequency distribution of inter-response time and of transition probability matrix in the amygdalectimized group. We thought that this impediment of DRL learning was due to the impediment of selection of decision of lever pressing action. The memory of the preamygdalectomy was maintained well after amygdalectomy.

Amygdala↗
PubMed1978

[Effect of blockers of serotonin and noradrenaline synthesis on learning with emotionally different reinforcement in rats].

The effect of inhibitor of serotonin and norepinephrine synthesis in the brain on learning was investigated in rats with emotionally different reinforcement. Parachlorphenylalanine (320 mg/kg) was shown to inhibit learning with food reinforcement, but facilitated learning with pain reinforcement. Disulfiram (100 mg/kg) inhibited learning with pain reinforcement considerably, but failed to influence learning with food reinforcement. Alpha-methyl-m-thyrosine inhibited both forms of learning. These new facts are in line with our previous data on mediating role of the brain monoaminergic systems between emotions and memory.

Animals↗
PubMed1979

Partial reinforcement in human biofeedback learning.

This paper reviews the evidence for the efficacy of partial reinforcement in producing resistance to extinction in human biofeedback experiments. The methodological criteria necessary to demonstrate such effects are discussed, as is the status of the analogy of reinforcement and information feedback. It is suggested that the problem of maintaining responding in the absence of feedback should be tackled empirically rather than assuming the validity of findings from other areas of learning theory.

Biofeedback, Psychology↗