SEARCH · Search PubMed

Results for “AI agents”

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.

SourceResearch connectionFrom yearThrough year

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

At least 19 recordsLinked to original sources

PubMed2026

Reimagining research papers as interactive and reliable AI agents.

Here we introduce Paper2Agent, an automated framework that converts research papers into artificial intelligence (AI) agents. Paper2Agent transforms research output from passive artefacts into active systems that accelerate use and discovery. Conventional research papers require readers to understand and adapt the paper's code, data and methods to their work, creating barriers to dissemination and reuse. Paper2Agent addresses this challenge by converting a paper into an AI agent that functions as a virtual corresponding author, exposing its manuscript, supplementary materials, datasets, code and workflows as active, agent-native knowledge rather than static text. It analyses the paper and codebase using multiple agents to construct a model context protocol (MCP) server, then generates and runs tests to refine and increase robustness of the MCP. These paper MCPs can be connected to a chat agent (such as Claude Code) to carry out complex scientific queries through natural language while invoking tools and workflows from the paper. We demonstrate Paper2Agent's effectiveness through case studies. Paper2Agent created an agent that leveraged AlphaGenome1 to interpret genomic variants and agents based on Scanpy2 and TISSUE (transcript imputation with spatial single-cell uncertainty estimation)3 to conduct single-cell and spatial transcriptomics analyses. We validate that these agents reproduce the results of the original papers and carry out novel user queries. Paper2Agent created multiple agents that collaborate to prioritize a causal gene for psoriasis. By turning static papers into interactive AI agents, Paper2Agent introduces a paradigm for knowledge dissemination and a collaborative ecosystem of AI co-scientists.

Journal Article↗

PubMed2026

Agentic AI for Spatial Omics.

This highlight summarises recent advances in agentic artificial intelligence (AI) systems for spatial omics analysis. These systems are compared along two central tensions: autonomy versus accountability, and adaptability versus reproducibility. We argue that progress will depend not on maximising automation, but on defining where autonomy is appropriate.

Artificial Intelligence↗

PubMed2026

Trustworthy Agentic AI in Bioinformatics: From Workflow Automation to Traceable and Validated Biological Inference.

Agentic artificial intelligence is extending bioinformatics beyond conversational assistance by enabling systems to select tools, execute code, revise analytical plans, and interpret biological data. These capabilities may accelerate research, but they also redistribute decisions that determine whether biological conclusions are valid. We conducted a targeted, structured PubMed search in July 2026 and identified 11 peer-reviewed agentic bioinformatics systems for descriptive review based on predefined eligibility criteria for analytical decision-making, tool or code execution, iterative evaluation, or coordinated agent activity. The evidence base covered single-cell transcriptomics, microbial genomics, cancer genomics, and omics applications, together with methodological literature on reproducibility and biological validation. We examined how current systems report delegated authority, provenance, validation, evidence, abstention, and human oversight. Existing platforms implement safeguards such as sandboxed execution, restricted commands, interaction logs, evidence identifiers, automated checks, critic agents, quality scores, and expert assessment. However, published reports rarely provide a connected account linking the original biological question to samples, reference resources, analytical decisions, computational actions, statistical results, supporting evidence, validation outcomes, and final claims. We distinguish inherited bioinformatics errors, errors amplified through autonomous action, and emergent failures arising from memory, retrieval, tool interaction, or agent coordination. We further propose a multidimensional decision-rights profile, consequence-sensitive validation gates, and a claim-to-evidence provenance architecture organized through the Traceable History of Research Evidence, Agent Actions, and Decisions in Bioinformatics (THREAD-Bio) framework. Illustrative cases show that technically successful execution may still support misleading inference. Trustworthy agentic bioinformatics therefore requires claims to remain reconstructible, challengeable, validated, and proportionate to the evidence.

accountable autonomy↗

PubMed2026

scBaseCount: An AI agent-curated, standardized, auto-updated single-cell data repository.

Single-cell RNA sequencing has transformed cell biology by enabling precise transcriptomic measurements of individual cells. The Sequence Read Archive (SRA) is the largest public repository of sequencing reads, yet much of it remains underutilized due to unstandardized metadata. Here, we introduce scBaseCount, a database that leverages an AI agent to automate discovery and metadata extraction and standardize data processing. Built by mining all 10x Genomics datasets, scBaseCount is the largest public repository of single-cell gene expression data, comprising over 502 million cells across 27 organisms and 75 tissues. It offers an unbiased view of the data landscape within the SRA and enables the training of more performant computational models through access to broader phenotypic diversity. Uniform processing enables measurement of both intronic and exonic reads and non-coding gene expression and improves alignment across experiments. Moreover, scBaseCount provides a blueprint for how AI can be leveraged to autonomously curate biological data repositories.

Single-Cell Analysis↗

PubMed2026

ELISA (Embedding-Linked Interactive Single-cell Agent): an interpretable hybrid generative Artificial Intelligence agent for expression-grounded discovery in single-cell genomics.

Translating single-cell RNA sequencing (scRNA-seq) data into mechanistic biological hypotheses remains a critical bottleneck, as agentic AI systems lack direct access to transcriptomic representations while expression foundation models remain opaque to natural language. Here, we introduce ELISA (Embedding-Linked Interactive Single-cell Agent), an interpretable framework that unifies single-cell generative pretrained transformer expression embeddings with biomedical bidirectional encoder representations from transformers-based semantic retrieval and large-language model (LLM)-mediated interpretation for interactive single-cell discovery. An automatic query classifier routes inputs to gene marker scoring, semantic matching, or reciprocal rank fusion pipelines depending on whether the query is a gene signature, natural language concept, or mixture of both. Integrated analytical modules perform pathway activity scoring across 60+ gene sets, ligand-receptor interaction prediction using 280+ curated pairs, condition-aware comparative analysis, and cell-type proportion estimation, all operating directly on embedded data without access to the original count matrix. Benchmarked across six diverse scRNA-seq datasets spanning inflammatory lung disease, pediatric and adult cancers, organoid models, healthy tissue, and neurodevelopment, ELISA significantly outperforms CellWhisperer, a classical lexical retriever (BM25), and a random baseline in cell type retrieval (combined permutation test, $p < 2\times 10^{-5}$ for each), with particularly large gains on gene-signature queries (Cohen's $d = 5.98$ for mean reciprocal rank). ELISA replicates published biological findings (mean composite score 0.88), and generates candidate hypotheses through grounded LLM reasoning, bridging the gap between transcriptomic data exploration and biological discovery.

Generative Artificial Intelligence↗

PubMed2025

Out-of-the-box bioinformatics capabilities of large language models (LLMs).

Large Language Models (LLMs), AI agents and co-scientists promise to accelerate scientific discovery across fields ranging from chemistry to biology. Bioinformatics- the analysis of DNA, RNA and protein sequences plays a crucial role in biological research and is especially amenable to AI-driven automation given its computational nature. Here, we assess the bioinformatics capabilities of three popular general-purpose LLMs on a set of tasks covering basic analytical questions that include code writing and multi-step reasoning in the domain. Utilizing questions from Rosalind, a bioinformatics educational platform, we compare the performance of the LLMs vs. humans on 104 questions undertaken by 110 to 68,760 individuals globally. GPT-3.5 provided correct answers for 59/104 (58%) questions, while Llama-3-70B and GPT-4o answered 49/104 (47%) correctly. GPT-3.5 was the best performing in most categories, followed by Llama-3-70B and then GPT-4o. 71% of the questions were correctly answered by at least one LLM. The best performing categories included DNA analysis, while the worst performing were sequence alignment/comparative genomics and genome assembly. Overall, LLMs performance mirrored that of humans with lower performance in tasks in which humans had low performance and vice versa. However, LLMs also failed in some instances where most humans were correct and, in a few cases, LLMs excelled where most humans failed. To the best of our knowledge, this presents the first assessment of general purpose LLMs on basic bioinformatics tasks in distinct areas relative to the performance of hundreds to thousands of humans. LLMs provide correct answers to several questions that require use of biological knowledge, reasoning, statistical analysis and computer code.

Journal Article↗

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↗

PubMed2025

AI-HOPE: an AI-driven conversational agent for enhanced clinical and genomic data integration in precision medicine research.

MOTIVATION: The growing complexity of clinical cancer research has fueled a surge in demand for automated bioinformatics tools capable of integrating clinical and genomic data to accelerate discovery efforts. RESULTS: We present the Artificial Intelligence Agent for High-Optimization and Precision Medicine (AI-HOPE), an AI-driven system that enables domain experts to conduct integrative data analyses through natural language interactions. Powered by Large Language Models, AI-HOPE interprets user instructions, converts them into executable code, and autonomously analyzes locally stored data. It supports flexible association studies, subset comparisons, clinical prevalence assessments and survival analyses. In addition, AI-HOPE enables global variable scans to identify features significantly associated with a user-defined outcome, making a powerful and intuitive tool for advancing precision medicine research. Importantly, its closed-system design prevents clinical data leakage. To demonstrate its utility, AI-HOPE was applied to The Cancer Genome Atlas data to address two clinical questions. First, it identified significant enrichment of TP53 mutations in late-stage colorectal cancer compared to early-stage cases. Second, it uncovered a strong association between KRAS mutations and poorer progression-free survival in FOLFOX-treated patients. These findings align with established literature and demonstrate AI-HOPE's ability to generate meaningful insights independently, without prior assumptions. By removing programming barriers and simplifying complex analyses, AI-HOPE bridges the gap between data complexity and research needs. With its scalable and adaptable framework, AI-HOPE has the potential to support diverse biomedical research fields, driving innovation and efficiency in translational studies. AVAILABILITY AND IMPLEMENTATION: The AI-HOPE software and demonstration data is available at https://github.com/Velazquez-Villarreal-Lab/AI-HOPE.

Precision Medicine↗

PubMed2026

Artificial intelligence agents and agentic artificial intelligence applied to precision medicine.

Precision medicine seeks to individualise care by integrating multimodal biomedical data, yet most deployed clinical artificial intelligence (AI) remains assistive, providing predictions without managing workflows or adapting autonomously. Agentic AI, built on large language models (LLMs), has emerged as a paradigm characterised by autonomy, goal-directed reasoning, memory, planning and tool use. This review synthesises evidence on agentic AI and LLMs applied to precision medicine, encompassing drug discovery, genomics, oncology, rare disease diagnostics and clinical pharmacology. This review also examines architectural components, recent validation milestones and emerging challenges, including hallucination, sociodemographic bias and evolving regulatory frameworks across the FDA, the EU AI Act and the WHO.

agentic AI↗

PubMed2026

Pharmacogenomic and drug interactions risk in cardio-oncology: A precision medicine perspective for India.

Cardio-oncology patients may face complex treatment regimens due to the concurrent existence of cancer and cardiovascular disease, leading to a considerable polypharmacy burden. This significantly increases the prospect of drug-drug interactions (DDIs) and gene-drug interactions. The majority of these interactions arise from comparable pharmacokinetic and pharmacological pathways associated with drug transporters and cytochrome P450 enzymes. The significance of pharmacogenomics in tailored treatment strategies are emphasised by the fact that genetic variability enhances individual differences in drug response, safety, and efficacy. This narrative review focus on the effects of key genetic polymorphisms (e.g., DPYD, CYP2C19, and CYP2C9) on the metabolism and efficacy of commonly prescribed anticancer and cardiovascular medications such as fluoropyrimidines, clopidogrel, and warfarin. In addition it explore the role of pharmacogenomic variants on drug-drug interactions within the field of cardio-oncology. The study ultimately emphasizes the necessity of precision medicine in India to address the genetic diversity and underrepresentation in global genomic databases. The absence of pharmacogenomic testing, infrastructural deficiencies, financial constraints, and insufficient clinical integration hinder the widespread use of this technology in India. The Genome India Project and other national initiatives establish the foundation for pharmacogenomic-guided therapy. Utilizing genetic data, together with artificial intelligence-based predictive tools, for clinical decision-making may enhance medication safety and yield optimal outcomes in Indian cardio-oncology patients.

Humans↗

PubMed1992

Effects of lovastatin and gemfibrozil on high-density lipoprotein subfraction density and composition in patients with familial hypercholesterolemia.

The effects of gemfibrozil and lovastatin treatment on composition and hydrated density distribution of high-density lipoprotein (HDL) were studied in 21 patients with heterozygous familial hypercholesterolemia with the use of HDL density gradient ultracentrifugation. At baseline the patients with familial hypercholesterolemia had a markedly reduced or missing HDL2 subfraction and their HDL3 was more dense with reduced content of cholesteryl ester and increased content of triglyceride compared with HDL of control subjects with normal lipid values. Gemfibrozil and lovastatin caused primarily similar alterations in HDL components in HDL2 and HDL3 subfractions. Both agents increased apolipoprotein AI and apolipoprotein AII concentrations significantly in HDL2, whereas the apolipoprotein changes in HDL3 were relatively smaller. The difference between the effects of these two agents was related to the HDL lipid composition. Gemfibrozil increased the cholesterol concentrations of HDL2 and HDL3 (p less than 0.05 for both), and lovastatin caused significant increases in HDL2 (p less than 0.05) and HDL3 phospholipids (p less than 0.01). The observed similarity of qualitative alterations in HDL subfractions produced by these two agents in patients with familial hypercholesterolemia differs from those reported in other types of hyperlipidemia and is probably a consequence of the basic abnormalities in HDL that are characteristic of familial hypercholesterolemia.

Adult↗

PubMed2026

APOM-associated inflammation and apoptosis in stroke-exacerbated myocardial infarction: implications for brain-heart interactions.

Brain-heart syndrome (BHS) describes cardiac dysfunction secondary to central nervous system injury, with acute ischemic stroke (AIS) serving as a critical driver that exacerbates myocardial infarction (MI). This study aimed to elucidate the role of Apolipoprotein M (APOM) in stroke-aggravated MI and to explore its underlying systemic and molecular mechanisms. Clinical data were analyzed to evaluate the correlation between stroke and MI. A combined mouse model of middle cerebral artery occlusion (MCAO) and MI was established to assess neurological and cardiac injury. Quantitative proteomics and Weighted Gene Co-expression Network Analysis (WGCNA) were employed to screen key differentially expressed proteins. The role of APOM in myocardial injury was validated using APOM-knockout (KO) mice. Furthermore, nuclear-cytoplasmic fractionation, immunofluorescence, and Western blot were performed to investigate its effects on the Saa1 and NF-κB signaling, NLRP3-related inflammatory signaling pathway, and lipid metabolism pathways. Clinical analysis indicated that stroke is a significant risk factor for MI (OR = 4.5). In the mouse model, MCAO significantly exacerbated post-MI electrocardiographic abnormalities, myocardial inflammatory response, while elevating circulating levels of cTnT and IL-1β. Proteomics identified a significant downregulation of APOM in the heart, brain, and serum post-stroke, a trend consistent with observations in AIS patients. Further experiments revealed that APOM deficiency markedly worsened cardiac conduction disturbances, histological damage, and inflammatory responses in MI mice. Mechanistically, the loss of APOM upregulates the acute-phase protein Saa1, triggers NF-κB phosphorylation and nuclear translocation, and enhances inflammatory signaling related to inflammasomes, while simultaneously mediating cytokine release from cardiomyocytes. Concurrently, APOM deficiency led to a significant decrease in sphingosine-1-phosphate (S1P) and also caused myocardial lipid droplet accumulation and metabolite changes. Additionally, the loss of APOM increased the expression of D-dimer and fibrinogen family proteins. Our findings suggest that APOM is a potential cardioprotective agent post-AIS. Downregulation of APOM may exacerbate myocardial injury after MI by elevating Saa1 expression, activating the NF-κB pathway and the inflammasome-mediated signaling, and inducing lipid metabolic disorders and coagulation-associated alterations. APOM may represent a potential therapeutic target for the intervention of brain-heart syndrome.

Animals↗

PubMed1976

Prolongation of allograft survival by combination therapy with anti-thymocyte serum and anti-immunoglobulin.

Heterologous anti-immunoglobulin (AI) is a potent immunosuppressive agent which compares favorably to anti-thymocyte serum (ATS) in inhibiting E and EAC-rosette formation and diminishing the secondary antibody response to sheep erythrocytes and E. coli lipopolysaccharide. AI prolongs H-2 incompatible skin allograft survival than either antiserum used alone, without evidence of toxicity to recipient mice. AI does not cause significant immune complex glomerular or renovascular deposition and may act by interference with antigen binding.

Animals↗

PubMed1975

Spermatogenetic clones developing from repopulating stem cells surviving a high dose of an alkylating agent.

We investigated stem cell renewal and differentiation in 10- and 15-days-old spermatogonial clones developing in mouse seminiferous epithelium after an extremely large cell loss, inflicted by high doses of the alkylating agent Myleran. The spermatogonial clones arise from cells that resemble the Ais spermatogonia but have a larger nuclear diameter. In spite of their mitotic activity these 'repopulating stem cells' lie mainly isolated or in pairs. This explained by migration and differentiation. Migration appeared to occur at random in all directions along the basement membrane of the seminiferous tubule. After one or more divisions of the stem cells, a second type of cell appears, which is called the 'differentiating spermatogomium'. The time elapsing before this type of cell appears, depends on the dose of Myleran: the larger the dose the later differentiation starts. A relation could be demonstrated between the stage of the cycle of the seminiferous epithelium and the start of differentiation. Differentiating cells were found isolated or in groups of two, four, eight or sixteen cells. Hence we concluded that at least up to their fourth division differentiating cells divide synchronously without degenerations. Three types of division of repopulating stem cells were distinguished, producing (1) two repopulating stem cells, (2) one repopulating stem cell and one cell starting spermatogonial differentiation, or (3) two differentiating cells. Type 1 divisions were found most frequently.

Animals↗

PubMed2026

Whole genome sequence of a superbug-Escherichia coli strain KAB-AI-497 isolated from the vagina of a 20 year old pregnant woman with premature rupture of membrane (PROM) in a resource limited setting, Kabale Regional Referral Hospital, in Uganda.

OBJECTIVES: The objective of the study is to sequence the whole genome of multidrug resistant E. coli strain KAB-AI-497 that causes bacterial vaginosis and implicated in premature rupture of membrane in pregnant woman. DATA DESCRIPTION: The DNA of the E. coli strain KAB-AI-497 was extracted using the MagAttract HMW DNA Kit, and the extracted DNA was sequenced using an MGI DNBSEQ G99ARS platform. FastQC was used to perform quality control analysis and the reads were trimmed by Trimmomatic. De novo genome assembly was performed by SPAdes and it resulted to a draft assembled genome that has 5.1 Mb genome size, 153 contigs, and 50.5% GC content. Quality analysis of the assembled genome revealed it has 98.46% completeness and 0.97% contamination. The closest E. coli strain to this strain KAB-AI-497 in terms of similarity was Escherichia coli SMS-3-5 with an average nucleotide identity of 98.43% and genome coverage of 86.19%, which confirmed the species level identity of the strain. The assembled genome was annotated using the NCBI Prokaryotic Genome Annotation Pipeline which identified 4,726 protein coding genes in the strain genome. Furthermore, the annotation revealed the genome has resistant genes responsible for resistance against many antibiotic classes such as tetracycline, fluoroquinolone, and penicillin.

Escherichia coli↗

PubMed2026

A comprehensive review of AI innovations for tackling antimicrobial resistance.

Antimicrobial resistance (AMR) represents a major global public health concern, rendering available antimicrobials ineffective and leading to infections that are difficult to treat. Artificial intelligence (AI) has been increasingly applied across the AMR continuum, including resistance prediction, rapid diagnostics, new antimicrobial discovery, drug repurposing, antimicrobial surveillance, and clinical decision support. In this review, we aim to highlight recent developments in the use of artificial intelligence (AI) to address antimicrobial resistance (AMR). In addition, we review computational methods that help interpret genomic, phenomic, clinical, and epidemiological data to support the development of treatment strategies and novel antimicrobial agents. The key issues addressed include data quality, model interpretability, external validation, regulatory requirements, privacy, and fairness. While AI is not a complete solution to AMR, it can certainly strengthen the global AMR response by complementing key areas of AMR such as antimicrobial stewardship, infection prevention, laboratory diagnostics, and global surveillance.

Antimicrobial resistance (AMR)↗

PubMed1992

Apolipoprotein synthesis and secretion in Hep G2 cells: effects of monensin and cycloheximide.

Hep G2 cells were used to study the relationship between apolipoprotein synthesis and secretion, as revealed by their interaction with agents modulating these processes. Cycloheximide inhibited the secretion of both apolipoproteins (apo) AI and B, but the reduction in apo AI secretion was evident at earlier times. Monensin also inhibited secretion of apo AI and apo B, but only apo AI accumulated intracellularly. Pulse-chase studies showed that, at concentrations of monensin that had no effect on total protein synthesis, apo B synthesis was specifically inhibited. Triacylglycerol synthesis was inhibited to the same extent as apo B synthesis, but this preceded the latter inhibition and unlike apo B there was an accumulation of intracellular triglyceride. These results suggest that distinctive mechanisms modulate the synthesis and secretion of apo AI and apo B, and that apo B synthesis can be specifically inhibited by mechanisms that initially block triglyceride production.

Apolipoprotein A-I↗

PubMed2026

Artificial intelligence for anticancer drug discovery from natural products of macroalgae and sponges: A systematic review.

Marine natural products (MNPs) from macroalgae and marine sponges have inspired clinically important anticancer agents, including the cytarabine pharmacophore and the eribulin scaffold, while cyanobacterial dolastatin chemistry supplies the auristatin payloads of several marine-inspired antibody-drug conjugates (ADCs) such as brentuximab vedotin. Artificial intelligence (AI) methods, encompassing both classical machine learning (ML) with hand-engineered features and modern deep learning (DL) with many-layered neural networks, are increasingly supporting key decisions in natural-product anticancer drug discovery, including bioactivity prediction, target identification, absorption, distribution, metabolism, excretion and toxicity (ADMET) filtering, generative analogue design, and the selection of preclinical candidates. DL architectures relevant to this field include graph neural networks, transformer-based molecular generators, diffusion models for protein-ligand docking, and convolutional networks for mass spectrometry, while classical ML contributes interpretable fingerprint-based bioactivity models and molecular networking for dereplication. This review follows a systematic literature review methodology to organize the landscape of AI methods now applied to MNP anticancer discovery, distinguishing ML and DL approaches where relevant, situating them within the chemical context of macroalgal and sponge-derived oncology leads, and critically examining published case studies, including validation level (computational, in vitro, in vivo, clinical). The principal bottleneck for medical translation has shifted partly from algorithmic capability toward data infrastructure and experimental validation. Sparse, heterogeneous, and taxonomically biased bioactivity records limit what current models can learn and reduce the reliability of AI-prioritized candidates entering the preclinical pipeline. A roadmap is proposed that prioritizes open MNP-specific benchmarks, symbiont-aware modeling, and active learning loops with synthesizability and ADMET constraints. These AI workflows may accelerate the prioritization of marine-derived anticancer leads and support earlier, more evidence-based translational decisions in oncology drug development.

Biological Products↗