SEARCH · Search PubMed

Results for “Deep Eutectic Solvents”

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.

53 recordsLinked to original sources

PubMed2026

Evaluation of the effects of domestic tomato processing on biopesticide residue using natural deep eutectic solvents (NADES) extractions.

The present study evaluated the fate of fourteen botanical biopesticides in processed tomato samples. Various processing methods were employed, including washing, dehydration, and the preparation of juice and sauce. The extraction was performed using more sustainable techniques, aimed at minimizing the environmental impact of conventional organic solvents by substituting them with natural deep eutectic solvents (NADES). Solid-liquid extraction (SLE) and dispersive liquid-liquid microextraction with solidification of floating organic drop (DLLME-SFOD) were utilized for solid and liquid tomato samples, respectively. The NADES used was choline chloride:2,3-butanediol (ChClBt) at a 1:4 molar ratio for both techniques, resulting in recovery values ranging from 69.2 to 106.2% for SLE, and extraction efficiencies reaching up to 46.2% for DLLME-SFOD. The impact of these processes was evaluated employing the processing factor (PF), yielding PF values of less than 1 in all cases. Compounds as pyrethrins, azadirachtin, and rotenone persisted after processing, posing a potential consumer risk.

Solanum lycopersicum↗

PubMed2026

Vortex-assisted liquid-liquid microextraction based on natural deep eutectic solvents for the determination of pyrethroid pesticides in urine.

A novel, facile, and environmentally friendly analytical method was developed based on vortex-assisted liquid-liquid microextraction and high-performance liquid chromatography with diode-array detection for detecting pyrethroid pesticides (PPs) in urine. Natural deep eutectic solvents (NADESs) were prepared using plant essential oil-derived monoterpenoids (thymol, carvacrol, and menthol) combined with aromatic primary alcohols (benzyl alcohol, phenethyl alcohol, and phenylpropyl alcohol) as hydrogen bond donors and acceptors. These solvents served as environmentally benign extraction media, thereby avoiding the use of conventional volatile, toxic organic solvents. NADESs are naturally derived, easy to prepare, biodegradable, and environmentally friendly solvents. Hydrophobic and π-π interactions between the NADESs and PPs may contribute to enhancing the affinity of PPs toward the NADESs phase. Vortex technology, accelerating mass transfer between the sample and extractant phases, enables fast extraction of PPs. Under optimized conditions, the method achieved a low detection limit (0.002 mg L-1), satisfactory precision with relative standard deviations (0.3%-2.4%), and acceptable recovery (80.7%-86.2%). The method demonstrated excellent performance in urine analysis and was feasible as a facile and green strategy for monitoring the content of PPs in biological matrices and assessing exposure risk.

Liquid Phase Microextraction↗

PubMed2026

Association between prenatal exposure to tetrachloroethylene and adverse birth outcomes: Systematic review and meta-analysis.

BACKGROUND: Tetrachloroethylene (PCE) is a ubiquitous chlorinated solvent with documented placental transfer. Despite widespread environmental and occupational exposure, no prior systematic review has synthesized evidence on prenatal PCE exposure and adverse birth outcomes. METHODS: We conducted a systematic review and meta-analysis of observational studies. PubMed, Web of Science, PsycINFO, EMBASE, and CINAHL were searched from inception to July 13, 2026. Eligible studies reported associations between prenatal PCE exposure (drinking water or inhalation) and adverse birth outcomes. Study quality was assessed using the Newcastle-Ottawa Scale (NOS) and Agency for Healthcare Research and Quality (AHRQ) criteria. Random-effects meta-analyses were performed using risk ratios (RRs) with 95% confidence intervals (CIs), with Knapp-Hartung adjustments and Paule-Mandel τ2 estimation. RESULTS: Twenty one studies (1987-2023) met inclusion criteria. Prenatal PCE exposure was associated with spontaneous abortion (8 studies; RR = 1.28, 95% CI 1.00-1.63; I2 = 64.2%). Analyses of stillbirth, central nervous system defects, oral clefts, neural tube defects, preterm birth, low birthweight, and small-for-gestational-age (SGA) yielded positive but statistically non-significant pooled estimates. The certainty of evidence ranged from very low to low across outcomes (GRADE). CONCLUSIONS: Prenatal PCE exposure may be associated with spontaneous abortion, particularly at higher exposure levels, and with SGA. Findings support ongoing regulatory efforts to limit PCE in occupational and environmental settings, particularly for pregnant individuals. Future prospective studies with biological monitoring and confounder-adjusted designs are needed.

Tetrachloroethylene↗

PubMed2026

Analysis of deep learning techniques in computer-aided diagnosis for meniscus injuries: a systematic literature review.

Meniscus informatics is a growing subject of study in the healthcare industry. One of the major hindrances to the healthcare system's transformation is obtaining knowledge and meaningful information from complicated, high-dimensional and diverse sources. Modern biomedical research, for instance, has seen an increase in the use of complex, dissimilar, poorly documented, and generally unstructured electronic health records, imaging, sensor data and text, even after many current techniques have been used to extract more robust and useful elements from the data for analysis. New efficient standards for building end-to-end learning models from complex data are therefore needed. Therefore, the current study aims to examine the most recent research on the use of deep learning techniques for diagnosing meniscus tears and recommend creating comprehensive and meaningful interpretable structures that might benefit the healthcare industry. We also draw attention to shortcomings and the need for better technique development, and we provide new perspectives about this exciting new development in the field.

Humans↗

PubMed2026

Systematic evaluation of one-dimensional-to-two-dimensional near-infrared spectroscopy transformations with deep learning for quantifying coconut sap adulteration.

Near-infrared (NIR) spectroscopy have limitations when combined with deep learning (DL) algorithms because they rely on low-dimensional datasets. Therefore, we investigated the potential of transforming one-dimensional (1D) NIR spectra into two-dimensional (2D) spectrograms using synchronous and asynchronous techniques and the continuous wavelet transform (CWT) and their effectiveness by integrating with DL for detecting adulteration in coconut sap. NIR spectra (12,500-4000 cm-1) were collected from binary mixtures (0%-100%;w/w). The performance of all DL (convolutional neural networks-CNN, AlexNet and ResNet) models was compared with that of partial least squares (PLS). The models were ranked in the mentioned order based on their performances: 2D-CWT > 2D-asynchronous > 2D-synchronous > 1D/2D-PLS. The important features of the best model can be explained and visualized using gradient-weighted-class-activation-mapping. The findings highlight that the 1D-to-2D NIR data transformation combined with DL is a highly robust approach because it addresses the feature representation gap in NIR data and effectively captures the spatial-spectral correlations.

Spectroscopy, Near-Infrared↗

PubMed2026

Comparison of deep and nondeep hypothermia in thoracic and thoracoabdominal aortic surgery: A systematic review and meta-analysis.

OBJECTIVE: Deep hypothermic circulatory arrest (DHCA) remains a cornerstone technique for neuroprotection and end-organ preservation during ascending aorta and arch surgeries. However, its benefits and risks compared with non-DHCA strategies in thoracic and thoracoabdominal aortic aneurysm (TAAA) repair are uncertain owing to conflicting evidence and variable institutional practices. METHODS: A systematic review and meta-analysis was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analysis and Cochrane guidelines. PubMed, Embase, and Cochrane Library were searched for comparative studies evaluating DHCA and non-DHCA techniques for open thoracic and TAAA repair. Random-effects models were applied to calculate pooled effect estimates. Effect sizes were risk ratio (RR) for binary end points and mean difference for continuous end points, both with 95% confidence intervals. Statistical significance was set at P < .05. Between-study heterogeneity was estimated using the I2 statistic. Metaregression analyses were used to explore the sources of heterogeneity. RESULTS: Nine observational studies, including 1041 patients, were analyzed. DHCA use was associated with a significantly lower risk of spinal cord injury (RR, 0.44; P = .012) compared with non-DHCA. However, DHCA was also associated with prolonged postoperative ventilation time (RR, 1.34; P = .003). No significant differences were observed in overall mortality, length of hospital and intensive care unit stay, stroke, or renal complications. Metaregression identified patient age as a moderator of length of stay variability, with older cohorts demonstrating longer recovery periods. CONCLUSIONS: DHCA is associated with a lower risk of spinal cord injury during TAAA repair without increasing mortality or stroke risk, although it is associated with longer ventilation times.

Humans↗

PubMed2026

Choice of Anesthesia in Microelectrode Recording-guided Deep Brain Stimulation Surgery for Parkinson's Disease (CHAMPION): A Noninferiority Randomized Controlled Trial.

BACKGROUND: Deep brain stimulation for Parkinson's disease is often performed under conscious sedation or general anesthesia. However, anesthetic agents may influence intraoperative microelectrode recording, and the optimal anesthesia method for microelectrode recording remains unclear. This study compared general anesthesia and conscious sedation in preserving microelectrode recording signal intensity during deep brain stimulation. METHODS: In this prospective, noninferiority randomized controlled trial, patients with Parkinson's disease (United Kingdom Brain Bank criteria) undergoing elective bilateral surgery were randomized 1:1 to the conscious sedation or the general anesthesia group. During surgery, a desflurane anesthetic titrated against the quality of the electrophysiologic signal was applied in the general anesthesia group, whereas patients in the conscious sedation group received dexmedetomidine anesthesia. The primary outcome was the proportion of patients with high-quality microelectrode recording (normalized root mean square greater than 2.0), assessed postoperatively off-line. Secondary outcomes included operation and recording duration, 6-month clinical efficacy, and complication rates. RESULTS: Of 188 randomized patients (94 general anesthesia, 93 conscious sedation), desflurane anesthesia was noninferior for high normalized root mean square proportion (89.4% vs . 90.3%; difference, -0.96%; 95% CI, -9.62 to 7.70). The general anesthesia group had shorter operative time (difference, -9.07 min; 95% CI, -13.99 to -4.14; P < 0.001). At 6 months, changes in Unified Parkinson's Disease Rating Scale score (difference, -2.50; 95% CI, -7.20 to 2.20; P = 0.297), levodopa equivalent daily dose (difference, -58.4 mg; 95% CI, -133.56 to 16.75; P = 0.128), and complication rates (general anesthesia: 10.9% vs . conscious sedation: 8.9%; P = 0.655) were comparable between the groups. CONCLUSIONS: General anesthesia is noninferior to conscious sedation for microelectrode-guided subthalamic nucleus deep brain stimulation, providing equivalent signal intensity and clinical outcomes while improving procedural efficiency, supporting its use as a valid clinical option.

Humans↗

PubMed2026

Target Capture of Ancient Shell DNA Enables Phylogenetic Reconstruction of Deep-Sea Molluscs.

Target capture is widely used to enrich endogenous DNA from calcium phosphate skeletal material in vertebrates, but its performance on calcium carbonate hard parts widely produced by invertebrates remains poorly understood. Here, we compared DNA recovery from four fresh and 12 ancient (eight radiocarbon-dated to 1671-1135 years old before present) deep-sea vesicomyid clam shells, including species Archivesica marissinica, A. nanshaensis and A. okutanii, using whole-genome sequencing (WGS) or target capture of ultraconserved elements (UCEs). WGS achieved 16.65% on-target read recovery of UCEs from fresh soft tissue, but < 1% from shell specimens. By contrast, UCE capture in the same specimen increased on-target reads by up to 155-fold, reaching 29.84% in fresh shells and up to 72-fold, reaching 19.89% in ancient shells. Target capture of UCEs recovered 142-1001 loci per sample compared to 0-230 with WGS alone. Ancient shells of A. marissinica and A. okutanii, based on reads mapped with bwa-mem2 and bbmap, exhibited characteristic post-mortem DNA damage signals, with average 5'-end C-to-T misincorporation rates of 3.46% and 15.97%, respectively, exceeding the levels observed in fresh A. marissinica shells (maximum 1.24%). UCE-based phylogenetic reconstructions incorporating shell ancient DNA recovered two major clades within Pliocardiinae, consistent with published phylogenomic trees. Together, these findings demonstrate that target-capture enrichment enables effective recovery of highly degraded DNA from ancient mollusc shells and supports robust phylogenetic inference at the intrageneric scale, expanding the utility of shells-one of the most abundant invertebrate remains-for evolutionary, biogeographic and conservation studies.

Animals↗

PubMed2026

Data-centric, robust, and explainable multimodal deep learning for clinical decision support: A systematic review.

PURPOSE: Multimodal deep learning is increasingly proposed for clinical decision support (CDS) under a "data-centric" framing that prioritizes label quality, missing-modality robustness, distribution shift, calibration, and explainability. Prior reviews have examined multimodal medical AI, CDS, and data-centric methods separately, but none address their intersection. We mapped the modalities, fusion strategies, and data-centric and explainability techniques used in this recent literature, quantified how often each is implemented rather than merely mentioned, assessed deployment-relevant evidence (external validation, clinical-outcome measurement, equity), and formally appraised study-level risk of bias. METHODS: Following the PRISMA 2020 statement (PROSPERO CRD420261427815; registered retrospectively), we screened 150 records and included primary, clinical, multimodal studies that applied machine or deep learning to a decision-support task and reported at least one quantitative result. Two reviewers screened and extracted data with consensus adjudication. Each study was coded against pre-specified operational definitions, separating implemented or empirically evaluated techniques from those only mentioned. Study-level risk of bias was assessed with PROBAST + AI. Synthesis was narrative. RESULTS: Thirty-one studies met inclusion; 30 (97%) were published between 2024 and 2026, with a median of three modalities (range 2-6), most commonly structured EHR (71%) and imaging (39%). Data-centric techniques were frequently reported (74-84% across label-noise, distribution-shift, calibration, missing-modality and class-imbalance handling; equity 61%). However, external validation was reported in only 4/31 studies (13%), a clinical or provider outcome in 3/31 (10%), and no study reported routine deployment. Overall risk of bias was high in 27/31 studies (87%), driven by the analysis domain. CONCLUSION: Within this recent, self-selected slice of the field, technical robustness and explainability techniques are widely reported but rarely validated out-of-distribution or against clinical outcomes, and the underlying evidence is at high risk of bias. Progress requires external multi-site validation, clinical-outcome measurement, formal bias appraisal, and adherence to AI reporting standards (e.g., TRIPOD + AI) before deployment can be justified.

Deep Learning↗

PubMed2026

The Effect of Slow Deep Breathing Relaxation Exercise on Pain and Anxiety Levels During and Post-Chest Tube Removal After CABG.

Chest tube removal after coronary artery bypass graft is frequently reported by patients as stressful and painful, highlighting the need for effective nonpharmacological interventions. Slow deep breathing relaxation exercises (SDBREs) may serve as a simple nursing strategy to reduce discomfort. In this study, we aimed to evaluate the effect of SDBRE on pain and anxiety during and after chest tube removal following coronary artery bypass grafting in Nablus hospitals. An experimental design was used with 80 patients recruited from 2 hospitals. Participants were randomly assigned to either an intervention group (n = 40) that practiced SDBRE or a control group (n = 40) that received standard care. Data were collected through a self-administered questionnaire, the Numeric Pain Scale, and the Visual Anxiety Scale. Data collection occurred from March to October 2024. The intervention group reported significantly lower pain scores during removal (M: 5.325 vs 7.125, P < .001) and after removal (P < .001). Anxiety scores were significantly lower both during and after removal (P < .001). Pain correlated with duration, with more complex operations and prolonged chest tube insertion linked to higher scores. SDBRE significantly reduced pain and anxiety during and after chest tube removal, supporting its integration into routine postoperative nursing care.

Humans↗

PubMed2026

Artificial intelligence for dental caries detection: An umbrella review.

Artificial intelligence (AI) has been proposed as a tool to improve dental caries detection across imaging modalities; however, its clinical value remains uncertain. This umbrella review aimed to synthesize and critically appraise systematic reviews evaluating AI for caries detection and diagnosis. An umbrella review was conducted following PRIOR guidance (PROSPERO CRD420261340728). Searches were performed in MEDLINE, Embase, Scopus, Web of Science, and Google Scholar up to 15 March 2026. Methodological quality was assessed using AMSTAR 2, and overlap of primary studies was quantified using the corrected covered area (CCA). Seventeen systematic reviews were included, of which five reported diagnostic test accuracy meta-analyses using bivariate or HSROC models. Across these meta-analyses, pooled sensitivity ranged from 0.76 to 0.94 and specificity from 0.85 to 0.91. Most systems were based on deep learning models applied to bitewing radiographs and intraoral photographs. However, substantial heterogeneity was observed in imaging modalities, lesion thresholds, analytical tasks, and evaluation metrics. In addition, a high degree of overlap across reviews and recurrent methodological limitations, including reliance on retrospective datasets, limited external validation, and inconsistent reporting, substantially weaken the reliability of the evidence. Although AI models demonstrate high diagnostic performance under experimental conditions, current evidence does not support their use as stand-alone diagnostic tools. Their clinical applicability remains limited, and implementation should be restricted to decision-support contexts until robust prospective validation demonstrates meaningful impact on clinical decision-making and patient outcomes.

Dental Caries↗

PubMed2026

Assessment of Surgical Salvage Outcomes for Exposed Cranial Neuromodulating Devices.

OBJECTIVES: Implanted neuromodulating devices (NMDs) such as cochlear implants (CIs) and deep brain stimulators (DBSs) are commonly used in modern medicine. Rarely, complications arise post-operatively, including hardware exposure. Traditional teaching suggests that these devices require removal if exposed; however, surgical salvage is a high risk, high reward alternative. We review our single institution experience managing NMD exposure with surgical salvage. METHODS: Retrospective chart review was performed on individuals who had a NMD implanted and underwent an attempt at surgical salvage for exposure during the study period (January 01, 2021 through December 31, 2023). Study outcome success was defined as maintaining a functional NMD 1 year after salvage was attempted. Surgical techniques associated with successful salvage were compared. RESULTS: Nine of 729 patients (1.2%) implanted with NMDs experienced hardware exposure during this 2-year study period. Nine subjects were referred for NMD salvage; however, only 6 of 9 subjects (66.7%, CI = 3; DBS = 3) underwent NMD salvage attempts. Four of the subjects had successful salvage demonstrated successful salvage with a functioning NMD and without wound healing concerns 1 year after their salvage procedure. CONCLUSIONS: Classic teaching states that exposed NMDs require explantation. However, this approach necessarily imposes time without benefit from the NMD between explantation and reimplantation. Our experience demonstrates that surgical salvage can be a successful alternative for the majority (66.7%) of individuals.

Humans↗

PubMed2026

Diversity and population connectivity of members of the family Eunicidae inhabiting deep-water corals in the North Atlantic.

Eunicid polychaetes are often found in association with Cold Water Corals (CWCs), even establishing symbiotic relationships, such as those described between Desmophyllum pertusum and Eunice norvegica. While genetic connectivity of CWCs across the North Atlantic has been widely studied, little is known about their associated fauna in this regard. Here, we present a study combining a focused analysis of the genetic and genomic connectivity of E. norvegica with a regional assessment of the distribution and evolutionary relationships of three CWC-associated eunicid species from the Cantabrian Sea and the North of the United Kingdom (190-1,230 m depth). An integrative approach using genetic (16S, COI and 18S), morphological and ecological data allowed the identification of the eunicids studied, with new records of Eunice cf. nicidioformis and Leodice cf. antarctica in the Cantabrian Sea, as well as previously undocumented associations with CWC species. In addition, RADseq data contributed to the delimitation of the closely related species E. norvegica and Eunice philocorallia. Moreover, the genetic connectivity of E. norvegica was studied trough a RADseq (1,067 neutral SNPs) approach. Our results indicate a single panmictic population across approximately 2,000 km, suggesting that oceanographic currents facilitate passive dispersal of E. norvegica lecithotrophic larvae, aided by coral host stepping-stones. The connectivity patterns observed for E. norvegica mirror those of D. pertusum, on which the worm is ecologically dependent. Our study highlights the importance of using integrated genetic, morphological and ecological data to characterise and delineate understudied CWC-associated species and improve our understanding of their dispersal capabilities and genetic connectivity to inform future conservation recommendations.

Animals↗

PubMed2026

Effects of H-coil TMS on suicidality in major depression: A secondary analysis of data from a multisite randomized trial comparing accelerated to once-a-day stimulation.

Suicide is the 10th leading cause of death in US adults. Standard once-daily repetitive transcranial magnetic stimulation (rTMS) can reduce suicidal ideation. Yet, antidepressant and anti-suicidal effects often take several weeks to emerge, while rapid improvement is often required. Accelerated TMS has been proposed as a strategy to hasten therapeutic response. A recent FDA-regulated multicenter trial evaluated accelerated intermittent theta burst Deep TMS with the H1-coil versus standard high-frequency Deep TMS in MDD. Both groups demonstrated high remission and response rates for depression, with the accelerated protocol showing non-inferiority and a shorter time to remission. The goal of this exploratory secondary analysis was to evaluate the impact of these two H-coil TMS dosing paradigms on suicidal ideation. The Scale for Suicide Ideation (SSI), as well as suicidality items of HDRS, MADRS and CUDOS were collected and analyzed. On all scales, both accelerated and standard Deep TMS protocols were associated with meaningful reductions in suicidality. The accelerated protocol achieved a faster onset of improvement. Comparison between the timeline of improvement in suicidality and in overall depressive symptoms found a trend for faster improvement in suicidality, especially with the accelerated protocol. These findings highlight the importance of treatment frequency in determining time to clinical benefit and support the use of scalable accelerated protocols for patients requiring more rapid symptom relief.

Humans↗

PubMed2026

Non-destructive prediction of lead content in oilseed rape leaves by fluorescence hyperspectral technology based on neural network.

Based on fluorescence hyperspectral imaging (FHSI), this study targeted rapid, non-destructive quantification of lead (Pb) content in oilseed rape leaves treated with varying silicon (Si) concentrations, acquiring fluorescence spectra over the 484.43-1001.61 nm wavelength range. To optimize spectral data quality, preprocessing methods (Savitzky-Golay smoothing, first derivative, detrending) were comprehensively compared. Characteristic wavelengths were then selected via interval variable iterative shrinkage, which effectively compressed data dimensionality and reduced computational load. A hybrid SE-CL1DA model, fusing a 1D convolutional neural network, a long short-term memory network and SE attention mechanism was constructed, with Bayesian optimization tuning hyperparameters to boost stability. The BO-SE-CL1DA outperformed both traditional machine learning and insufficiently optimized deep learning model (Rp2=0.9609, RMSE = 0.0377 mg/kg, RPD = 5.1736), thus enabling accurate Pb estimation, supporting Si-regulated heavy metal stress management and facilitating agricultural contamination monitoring.

Plant Leaves↗

PubMed2026

Integrating genomic distance analyses in the description of a new family, genus, and species of sponge-associated antipatharians (black corals).

Antipatharians (black corals) are among the least studied coral groups, with much of their diversity still undescribed. Here, we present an integrative morphological, phylogenomic and genomic distance study of deep-sea antipatharians sampled in high seas areas of the North Pacific Ocean and from New Zealand's Exclusive Economic Zone. These corals grow on hexactinellid sponges - a unique characteristic in the order Antipatharia. Using a dataset of ultra-conserved elements and exons, combined with morphological analyses, we reconstruct phylogenomic relationships and formally describe a new family (Eidikopathidae fam. nov.), a new genus (Eidikopathesgen. nov.), and two new species (E. korallispongiasp. nov., E. zealandkoralliasp. nov.). Morphologically, the new family is distinguished by a corallum consisting of a network of loose branches that fuse with the sponge skeletal framework. Phylogenomic analyses recovered consistent topologies with strong nodal support, corroborating the distinct evolutionary placement of this sponge-associated lineage. Pairwise genomic distances estimated using the Tamura-Nei model were concordant with patristic genomic distances, identifying Pteridopathidae as the genetically closest family to Eidikopathidae fam. nov., followed by Myriopathidae and Stylopathidae, which were recovered as sister families in the phylogeny. This pattern shows that genomic distance complements, rather than simply mirrors, tree topology by quantifying accumulated sequence divergence among lineages. Together, these results provide the first genomic distance framework for Antipatharia, offering a baseline for future systematic, evolutionary, and biodiversity studies on this fundamental shallow, mesophotic and deep-sea coral group.

Animals↗

PubMed2026

Meta-PseU: A meta-classifier for robust prediction of RNA pseudouridine modification sites from long sequences.

BACKGROUND AND OBJECTIVES: Pseudouridine (Ψ) represents one of the most abundant and conserved RNA modifications. Ψ provides an additional hydrogen-bond donor that enhances RNA structural stability and modulates translation. It participates in diverse biological processes, including RNA-protein interactions, splicing, translational control, and stress responses. Aberrant pseudouridylation is implicated in cancer, neurodegenerative disorders, and autoimmune diseases. Despite its biological importance, experimental identification of Ψ sites remains time-consuming and costly, limiting the feasibility of transcriptome-wide profiling. Computational approaches have therefore become essential complements to experimental techniques. However, state-of-the-art machine-learning and deep-learning predictors often suffer from limited generalizability due to small training datasets. To overcome these issues, we aim at constructing new long-sequence datasets and developing a novel Ψ site predictor. METHODS: New long-sequence datasets were constructed as benchmarks for RNA Ψ-site prediction. The Ψ modification sites in RMBase 3.0 were mapped to the reference genomes across three species of human, mouse, and yeast, and the RNA sequences with a length of 201 were generated by extending the upstream and downstream from the mapped, central sites. To eliminate sequence redundancy, the sequences were clustered using CD-HIT with a 70% sequence identity threshold. We developed Meta-PseU, a logistic regression-based meta-classifier that considered 118 machine learning and deep learning classifiers. The datasets and programs are freely accessible at https://github.com/kuratahiroyuki/MetaPseU. RESULTS: By optimizing model configuration, we proposed the Meta-PseU model stacking 32 machine learning and deep learning classifiers out of 118 classifiers. Meta-PseU substantially improved model generalizability, overcoming a key limitation of existing approaches. It greatly outperformed state-of-the-art predictors and achieved increasing accuracy with increasing sequence length. CONCLUSIONS: Long-sequence datasets were newly constructed as benchmarks for RNA Ψ-site prediction. Meta-PseU offers a new framework for robust Ψ-site identification by using long sequences.

Pseudouridine↗

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↗