BMC Genomics. 2025 Feb 4;26(1):106. doi: 10.1186/s12864-025-11308-3.

ABSTRACT

BACKGROUND: Chickpea (Cicer arietinum L.) ranks as the third most crucial grain legume worldwide. Fusarium wilt (Fusarium oxysporum f. sp. ciceri (Foc)) is a devastating fungal disease that prevents the maximum potential for chickpea production.

RESULTS: To identify genes and pathways involved in resistance to race 6 of Foc, this study utilized transcriptome sequencing of two chickpea cultivars: resistant (Ana) and susceptible (Hashem) to Foc race 6. Illumina sequencing of the root samples yielded 133.5 million raw reads, with about 90% of the clean reads mapped to the chickpea reference genome. The analysis revealed that 548 genes (332 upregulated and 216 downregulated) in the resistant genotype (Ana) and 1115 genes (595 upregulated and 520 downregulated) in the susceptible genotype (Hashem) were differentially expressed under Fusarium wilt (FW) disease stress caused by Foc race 6. The expression patterns of some differentially expressed genes (DEGs) were validated using quantitative real-time PCR. A total of 131 genes were exclusively upregulated under FW stress in the resistant cultivar, including several genes involved in sensing (e.g., CaNLR-RPM1, CaLYK5-RLK, CaPR5-RLK, CaLRR-RLK, and CaRLP-EIX2), signaling (e.g., CaPP7, CaEPS1, CaSTY13, and CaPR-1), transcription regulation (e.g., CaMYBs, CaGLK, CaERFs, CaZAT11-like, and CaNAC6) and cell wall integrity (e.g., CaPGI2-like, CaEXLs, CaCSLD and CaCYP73A100-like).

CONCLUSIONS: The achieved results could provide insights into the molecular mechanism underlying resistance to FW and could be valuable for breeding programs aimed at developing FW-resistant chickpea varieties.

PMID:39905311 | DOI:10.1186/s12864-025-11308-3

Genet Sel Evol. 2025 Feb 4;57(1):3. doi: 10.1186/s12711-025-00951-9.

ABSTRACT

BACKGROUND: Milk production traits are complex and influenced by many genetic and environmental factors. Although extensive research has been performed for these traits, with many associations unveiled thus far, due to their crucial economic importance, complex genetic architecture, and the fact that causal variants in cattle are still scarce, there is a need for a better understanding of their genetic background. In this study, we aimed to identify new candidate loci associated with milk production traits in German Holstein cattle, the most important dairy breed in Germany and worldwide. For that purpose, 180,217 cattle were imputed to the sequence level and large-scale genome-wide association study (GWAS) followed by fine-mapping and evolutionary and functional annotation were carried out to identify and prioritize new association signals.

RESULTS: Using the imputed sequence data of a large cattle dataset, we identified 50,876 significant variants, confirming many known and identifying previously unreported candidate variants for milk (MY), fat (FY), and protein yield (PY). Genome-wide significant signals were fine-mapped with the Bayesian approach that determines the credible variant sets and generates the probability of causality for each signal. The variants with the highest probabilities of being causal were further classified using external information about the function and evolution, making the prioritization for subsequent validation experiments easier. The top potential causal variants determined with fine-mapping explained a large percentage of genetic variance compared to random ones; 178 variants explained 11.5%, 104 explained 7.7%, and 68 variants explained 3.9% of the variance for MY, FY, and PY, respectively, demonstrating the potential for causality.

CONCLUSIONS: Our findings proved the power of large samples and sequence-based GWAS in detecting new association signals. In order to fully exploit the power of GWAS, one should aim at very large samples combined with whole-genome sequence data. These can also come with both computational and time burdens, as presented in our study. Although milk production traits in cattle are comprehensively investigated, the genetic background of these traits is still not fully understood, with the potential for many new associations to be revealed, as shown. With constantly growing sample sizes, we expect more insights into the genetic architecture of milk production traits in the future.

PMID:39905301 | DOI:10.1186/s12711-025-00951-9

Nat Genet. 2025 Feb 4. doi: 10.1038/s41588-025-02078-5. Online ahead of print.

ABSTRACT

Cell division drives somatic evolution but is challenging to quantify. We developed a framework to count cell divisions with DNA replication-related mutations in polyguanine homopolymers. Analyzing 505 samples from 37 patients, we studied the milestones of colorectal cancer evolution. Primary tumors diversify at ~250 divisions from the founder cell, while distant metastasis divergence occurs significantly later, at ~500 divisions. Notably, distant but not lymph node metastases originate from primary tumor regions that have undergone surplus divisions, tying subclonal expansion to metastatic capacity. Then, we analyzed a cohort of 73 multifocal lung cancers and showed that the cell division burden of the tumors' common ancestor distinguishes independent primary tumors from intrapulmonary metastases and correlates with patient survival. In lung cancer too, metastatic capacity is tied to more extensive proliferation. The cell division history of human cancers is easily accessible using our simple framework and contains valuable biological and clinical information.

PMID:39905260 | DOI:10.1038/s41588-025-02078-5

Nat Immunol. 2025 Feb 4. doi: 10.1038/s41590-024-02075-6. Online ahead of print.

ABSTRACT

Regulatory T (Treg) cells are a specialized CD4+ T cell lineage with essential anti-inflammatory functions. Analysis of Treg cell adaptations to non-lymphoid tissues that enable their specialized immunosuppressive and tissue-supportive functions raises questions about the underlying mechanisms of these adaptations and whether they represent stable differentiation or reversible activation states. Here, we characterize distinct colonic effector Treg cell transcriptional programs. Attenuated T cell receptor (TCR) signaling and acquisition of substantial TCR-independent functionality seems to facilitate the terminal differentiation of a population of colonic effector Treg cells that are distinguished by stable expression of the immunomodulatory cytokine IL-10. Functional studies show that this subset of effector Treg cells, but not their expression of IL-10, is indispensable for colonic health. These findings identify core features of the terminal differentiation of effector Treg cells in non-lymphoid tissues and their function.

PMID:39905200 | DOI:10.1038/s41590-024-02075-6

Nat Rev Gastroenterol Hepatol. 2025 Feb 4. doi: 10.1038/s41575-025-01039-x. Online ahead of print.

NO ABSTRACT

PMID:39905174 | DOI:10.1038/s41575-025-01039-x

Nat Commun. 2025 Feb 4;16(1):1346. doi: 10.1038/s41467-025-56590-7.

ABSTRACT

Cell signaling plays a critical role in neurodevelopment, regulating cellular behavior and fate. While multimodal single-cell sequencing technologies are rapidly advancing, scalable and flexible profiling of cell signaling states alongside other molecular modalities remains challenging. Here we present Phospho-seq, an integrated approach that aims to quantify cytoplasmic and nuclear proteins, including those with post-translational modifications, and to connect their activity with cis-regulatory elements and transcriptional targets. We utilize a simplified benchtop antibody conjugation method to create large custom neuro-focused antibody panels for simultaneous protein and scATAC-seq profiling on whole cells, alongside both experimental and computational strategies to incorporate transcriptomic measurements. We apply our workflow to cell lines, induced pluripotent stem cells, and months-old retinal and brain organoids to demonstrate its broad applicability. We show that Phospho-seq can provide insights into cellular states and trajectories, shed light on gene regulatory relationships, and help explore the causes and effects of diverse cell signaling in neurodevelopment.

PMID:39905064 | DOI:10.1038/s41467-025-56590-7

Nat Commun. 2025 Feb 4;16(1):1347. doi: 10.1038/s41467-025-56646-8.

ABSTRACT

Ion suppression is a major problem in mass spectrometry (MS)-based metabolomics; it can dramatically decrease measurement accuracy, precision, and sensitivity. Here we report a method, the IROA TruQuant Workflow, that uses a stable isotope-labeled internal standard (IROA-IS) library plus companion algorithms to: 1) measure and correct for ion suppression, and 2) perform Dual MSTUS normalization of MS metabolomic data. We evaluate the method across ion chromatography (IC), hydrophilic interaction liquid chromatography (HILIC), and reversed-phase liquid chromatography (RPLC)-MS systems in both positive and negative ionization modes, with clean and unclean ion sources, and across different biological matrices. Across the broad range of conditions tested, all detected metabolites exhibit ion suppression ranging from 1% to >90% and coefficients of variation ranging from 1% to 20%, but the Workflow and companion algorithms are highly effective at nulling out that suppression and error. To demonstrate a routine application of the Workflow, we employ the Workflow to study ovarian cancer cell response to the enzyme-drug L-asparaginase (ASNase). The IROA-normalized data reveal significant alterations in peptide metabolism, which have not been reported previously. Overall, the Workflow corrects ion suppression across diverse analytical conditions and produces robust normalization of non-targeted metabolomic data.

PMID:39905052 | DOI:10.1038/s41467-025-56646-8

Nat Commun. 2025 Feb 4;16(1):1356. doi: 10.1038/s41467-025-56533-2.

ABSTRACT

Pluripotent stem cells provide a scalable approach to analyse molecular regulation of cell differentiation across developmental lineages. Here, we engineer barcoded induced pluripotent stem cells to generate an atlas of multilineage differentiation from pluripotency, encompassing an eight-day time course with modulation of WNT, BMP, and VEGF signalling pathways. Annotation of in vitro cell types with reference to in vivo development reveals diverse mesendoderm lineage cell types including lateral plate and paraxial mesoderm, neural crest, and primitive gut. Interrogation of temporal and signalling-specific gene expression in this atlas, evaluated against cell type-specific gene expression in human complex trait data highlights the WNT-inhibitor gene TMEM88 as a regulator of mesendodermal lineages influencing cardiovascular and anthropometric traits. Genetic TMEM88 loss of function models show impaired differentiation of endodermal and mesodermal derivatives in vitro and dysregulated arterial blood pressure in vivo. Together, this study provides an atlas of multilineage stem cell differentiation and analysis pipelines to dissect genetic determinants of mammalian developmental physiology.

PMID:39904980 | DOI:10.1038/s41467-025-56533-2

Nat Commun. 2025 Feb 4;16(1):1094. doi: 10.1038/s41467-024-55085-1.

ABSTRACT

UDP-GlcNAc serves as a building block for glycosaminoglycan (GAG) chains in cartilage proteoglycans and simultaneously acts as a substrate for O-GlcNAcylation. Here, we show that transporters for UDP-GlcNAc to the endoplasmic reticulum (ER) and Golgi are significantly downregulated in osteoarthritic cartilage, leading to increased cytosolic UDP-GlcNAc and O-GlcNAcylation in chondrocytes. Mechanistically, upregulated O-GlcNAcylation governs the senescence-associated secretory phenotype (SASP) by stabilizing GATA4 via O-GlcNAcylation at S406, which compromises its degradation by p62-mediated selective autophagy. Elevated O-GlcNAcylation in the superficial layer of osteoarthritic cartilage coincides with increased GATA4 levels. The topical deletion of Gata4 in this cartilage layer ameliorates post-traumatic osteoarthritis (OA) in mice while inhibiting O-GlcNAc transferase mitigates OA by decreasing GATA4 levels. Excessive glucosamine-induced O-GlcNAcylation stabilizes GATA4 in chondrocytes and exacerbates post-traumatic OA in mice. Our findings elucidate the role of UDP-GlcNAc compartmentalization in regulating secretory pathways associated with chronic joint inflammation, providing a senostatic strategy for the treatment of OA.

PMID:39904978 | DOI:10.1038/s41467-024-55085-1

Front Immunol. 2025 Jan 21;15:1508129. doi: 10.3389/fimmu.2024.1508129. eCollection 2024.

ABSTRACT

QUESTION: Can previously reported, largely anecdotal associations between exposure to any of a comprehensive list of putative trigger drugs and the development of pemphigus be reproduced using population level data?

FINDINGS: In this series of observational, retrospective, case-control, pharmacovigilance analyses of the FDA Adverse Event Reporting System, the odds of reporting the adverse event pemphigus were significantly elevated among individuals exposed to 11/36 previously reported trigger drugs namely, gold sodium thiomalate, penicillamine, piroxicam, rifampin, hydroxychloroquine, imiquimod, hydrochlorothiazide, irbesartan, lisinopril, nivolumab, and nifedipine.

MEANING: Environmental exposures such as drugs are relevant players in the pathogenesis of autoimmune diseases and clinicians who treat patients with autoimmune blistering diseases such as pemphigus should consider performing a detailed medication history leveraging this information regarding deleterious drug-disease interactions at initial evaluation as well as longitudinal monitoring of patients to better inform clinical care decisions.

IMPORTANCE: Pemphigus vulgaris (PV) is a rare, potentially fatal autoimmune disease with pathogenic contributions from both genetic as well as environmental factors, notably drug exposures. Despite anecdotal reports linking multiple drugs to PV, corroborating evidence from large datasets is missing.

OBJECTIVE: To examine the extent to which previously reported associations between a comprehensive list of 36 drugs implicated in PV pathogenesis could be replicated using population-level pharmacovigilance data.

DESIGN: Series of observational, retrospective, case-control, pharmacovigilance analyses (one analysis/drug, 36 total).

SETTING: Population based.

PARTICIPANTS: Individuals who submitted a report of a drug-related adverse event to the FDA from Q4 of 2003 to Q2 of 2023.

EXPOSURE: Cases were identified by the presence of adverse events described by the MedDRA preferred term "pemphigus" (10034280) and then sorted based on exposure to each of the drugs of interest.

MAIN OUTCOMES AND MEASURES: Reporting Odds Ratios (RORs) quantifying the association between a given drug exposure and reports of pemphigus adverse events.

RESULTS: The analyses revealed statistically significant associations between reports of pemphigus and exposure to 11/36 previously reported drugs, two of which had particularly high RORs (>200) [gold sodium thiomalate (ROR, 266.0; 95% CI, 202.6-349.3) and hydroxychloroquine (ROR, 282.6; 95% CI, 261.0-306.1)], three had very strong RORs (14-45) [penicillamine (ROR, 30.5; 95% CI, 11.4-81.7), piroxicam (ROR, 14.8; 95% CI, 8.2-26.7), and imiquimod (ROR, 42.3; 95% CI, 26.2-68.3)], and six had modestly strong RORs (2-5) [rifampin (ROR, 2.8; 95% CI, 1.4-5.6), hydrochlorothiazide (ROR, 1.6; 95% CI, 1.2-2.1), irbesartan (ROR, 2.7; 95% CI, 1.6-4.4), lisinopril (ROR, 5.3; 95% CI, 4.5-6.2), nivolumab (ROR, 2.7; 95% CI, 1.8-4.1), and nifedipine (ROR, 3.0; 95% CI, 1.9-5.0)]. Associations for other previously reported drugs (25/36) were not detected.

CONCLUSIONS AND RELEVANCE: This study represents a comprehensive evaluation of suspected drug triggers of pemphigus using real-world data. The significant associations reported here provide empirical support for the hypothesis that certain drugs act as triggers for PV. Moreover, all of the drugs found to be associated with PV in this study harbor immunomodulatory capacity, suggesting that the ability to induce such perturbations, directly or indirectly, may be a critical factor connecting drug exposure to pemphigus pathogenesis. However, the absence of signals for other previously reported putative trigger drugs does not preclude their potential role in PV pathogenesis. Our findings reinforce the need for larger, more definitive studies to confirm these associations and to explore the mechanisms by which these drugs may contribute to PV development. Finally, these findings underscore the importance of considering environmental factors in the development and course of PV in genetically susceptible individuals.

PMID:39906745 | PMC:PMC11790476 | DOI:10.3389/fimmu.2024.1508129

Clin Transl Sci. 2025 Feb;18(2):e70133. doi: 10.1111/cts.70133.

ABSTRACT

Piperaquine tetraphosphate (PQP), a long-acting antimalarial, is being considered in a combination for chemoprevention. Dihydroartemisinin-piperaquine tablets (hard and dispersible) approved for the treatment of acute uncomplicated malaria should be administered in a fasted state, as PQP bioavailability increases with food. A new taste-masked dispersible granules PQP formulation aims to minimize the impact of food on drug exposure. This randomized, open label, parallel group, Phase I pilot study was conducted between 24th July 2023, and 3rd January 2024, at the Ifakara Health Institute, Bagamoyo, Tanzania in 60 healthy African adults (five cohorts of 12). Single-dose pharmacokinetics and relative systemic exposure of the oral PQP dispersible granules formulation prototype (320 mg) was compared to the hard tablet when fasted and PQP granules in three different fed conditions. In the fasted state, the relative exposure of PQP granules versus the tablet was 73.9% (90% CI 48.3, 113.0) for Cmax and 86.5% (68.2, 109.6) for AUC0-t. Following a typical East African low-fat meal, a standard high-fat meal, or 250 mL whole milk, the relative exposure of PQP granules versus the fasted state was 202% (90% CI 132, 311), 275% (193, 391), and 294% (203, 425) for Cmax and 164% (124, 217), 184% (148, 228), and 195% (147, 259) for AUC0-t, respectively. Both formulations were well tolerated with one drug-related adverse event (moderate migraine). No severe or serious adverse events or clinically relevant laboratory or electrocardiograph changes were observed. PQP dispersible granules had lower systemic exposures versus the tablet when fasted, whereas various meals increased drug exposure.

PMID:39905737 | DOI:10.1111/cts.70133

Sci Rep. 2025 Feb 4;15(1):4157. doi: 10.1038/s41598-025-87724-y.

ABSTRACT

Adverse Drug Reactions (ADRs) stand out as a pressing challenge in public health and a critical aspect of drug discovery. The dilemma arises from the inherent impossibility of conducting a comprehensive evaluation of a drug before its market release, constrained by the limitations in scale and duration of clinical trials. Therefore, the post-marketing detection of ADRs in a timely and accurate manner becomes imperative. Adding to the complexity, a multitude of tweets harbor concealed information about adverse drug reactions, creating difficulties due to their concise, sporadic, and noisy content. To solve the problem, we regard ADR detection as a question-answer problem and introduces an innovative neural network framework with multiple GRU layers designed for extracting ADR-related information from tweets. The Von Mises-Fisher distribution is applied to derive keyword vectors through tweet sampling. An attention mechanism is employed to enhance the interaction between these keyword vectors and the word sequences within tweets. The credibility of word sequences is systematically evaluated based on the reliability of answer factors. To address concerns related to background information and training speed, we propose a quality assurance mechanism utilizing a GRU network due to its straightforward structure and efficient training capabilities. As a result of the training process, word sequences are mapped to a low-latitude vector space, generating corresponding answers. Experimental results obtained from two Twitter ADR datasets affirm that our Question-Answer Mechanism, leveraging multi-GRU architecture, significantly improves the accuracy of ADR detection in tweets. Our method achieved F1-scores of 81.3% and 73.3% on the two datasets, respectively, while consistently maintaining a higher recall.

PMID:39905141 | DOI:10.1038/s41598-025-87724-y

J Imaging Inform Med. 2025 Feb 4. doi: 10.1007/s10278-025-01428-3. Online ahead of print.

ABSTRACT

Vaginitis is a common condition in women that is described medically as irritation and/or inflammation of the vagina; it poses a significant health risk for women, necessitating precise diagnostic methods. Presently, conventional techniques for examining vaginal discharge involve the use of wet mounts and gram staining to identify vaginal diseases. In this research, we utilized fluorescent staining, which enables distinct visualization of cellular and pathogenic components, each exhibiting unique color characteristics when exposed to the same light source. We established a large, challenging multiple fluorescence leucorrhea dataset benchmark comprising 8 categories with a total of 343 K high-quality labels. We also presented a robust lightweight deep-learning network, LRNet. It includes a lightweight feature extraction network that employs Ghost modules, a feature pyramid network that incorporates deformable convolution in the neck, and a single detection head. The evaluation results indicate that this detection network surpasses conventional networks and can cut down the model parameters by up to 91.4% and floating-point operations (FLOPs) by 74%. The deep-optimal leucorrhea detection capability of LRNet significantly enhances its ability to detect various crucial indicators related to vaginal health.

PMID:39904942 | DOI:10.1007/s10278-025-01428-3

J Imaging Inform Med. 2025 Feb 4. doi: 10.1007/s10278-024-01376-4. Online ahead of print.

ABSTRACT

The study aimed to predict expression of pituitary transcription factor 1 (PIT1) in pituitary adenomas using habitat, intra-tumoral and peri-tumoral radiomics models. A total of 129 patients with pituitary adenoma (training set, n = 103; test set, n = 26) were retrospectively enrolled. A total of 12, 18, 14, 13, and 14 radiomics features were selected from the ROIintra, ROIintra+peri (ROIintra+2mm, ROIintra+4mm, ROIintra+6mm), and ROIhabitat, respectively. Then, three machine learning algorithms were employed to develop radiomic models, including logistic regression (LR), support vector machines (SVM), and multilayer perceptron (MLP). The performances of the intra-tumoral, combined intra-tumoral and peri-tumoral, and habitat models were evaluated. The peritumoral region (ROI2mm, ROI4mm, ROI6mm) of the combined model with the highest performance was individually selected for further peritumoral analysis. Moreover, a deep learning radiomics nomogram (DLRN) was constructed incorporating clinical characteristics and the peri-tumoral and habitat models for individual prediction. The combined modelintra+2mm based on ROIintra+2mm achieved a better performance (AUC, 0.800) than that of the intra-tumoral model alone (AUC, 0.731). And the habitat model showed a higher performance (AUC, 0.806) than that of the intra-tumoral model. In addition, the performance of the peri-tumoral model based on ROI2mm was 0.694 in the testing set. Furthermore, the DLRN achieved the highest performance of 0.900 in the test set. The DLRN showed the best performance for PIT1 expression in pituitary adenomas, followed by the habitat, combined modelintra+2mm, intra-tumoral model, and peri-tumoral model based on ROI2mm, respectively. These different models are helpful for the model choice in clinical work.

PMID:39904941 | DOI:10.1007/s10278-024-01376-4

J Imaging Inform Med. 2025 Feb 4. doi: 10.1007/s10278-025-01419-4. Online ahead of print.

ABSTRACT

Diabetic retinopathy is a major complication of diabetes, with its prevalence nearly doubling to approximately 10.5% by 2021. Exudates, the characteristic lesions of diabetic retinopathy, are crucial for assessing disease progression and severity. The location and distribution of these exudates can affect various regions of the retina, necessitating a detailed regional analysis of lesions. To address this need, this study aimed to evaluate the performance of exudate detection in fundus images across various regions, including perivascular and extravascular areas, perifoveal and extrafoveal regions, and in quadrants defined relative to the fovea. We employed U-net and U-net3 + deep learning models for validation, evaluating their performance using accuracy, sensitivity, specificity, and Dice score. Overall, the U-net3 + model outperformed the U-net model. Therefore, the performance evaluation was based on the results from the U-net3 + model. Comparing the detection performance across perivascular versus extravascular and perifoveal versus extrafoveal regions, the U-net3 + model achieved highest Dice score in the extravascular (87.96% [± 5.80]) and perifoveal areas (88.03% [± 5.86]). Additionally, superior sensitivity and Dice scores were observed in the top-left and top-right quadrants. Future research is anticipated to show that deep learning-based automatic exudate detection will enhance diagnostic accuracy and efficiency, leading to better treatment and prognosis in patients with diabetic retinopathy.

PMID:39904940 | DOI:10.1007/s10278-025-01419-4

Respir Investig. 2025 Feb 3;63(2):224-225. doi: 10.1016/j.resinv.2025.01.002. Online ahead of print.

NO ABSTRACT

PMID:39904248 | DOI:10.1016/j.resinv.2025.01.002

Neurocrit Care. 2025 Feb 4. doi: 10.1007/s12028-025-02217-0. Online ahead of print.

ABSTRACT

BACKGROUND: Despite advancements in the neuroscience of consciousness, no new medications for disorders of consciousness (DOC) have been discovered in more than a decade. Repurposing existing US Food and Drug Administration (FDA)-approved drugs for DOC is crucial for improving clinical management and patient outcomes.

METHODS: To identify potential new treatments among existing FDA-approved drugs, we used a deep learning-based drug screening model to predict the efficacy of drugs as awakening agents based on their three-dimensional molecular structure. A retrospective cohort study from March 2012 to October 2024 tested the model's predictions, focusing on changes in Glasgow Coma Scale (GCS) scores in 4047 patients in a coma from traumatic, vascular, or anoxic brain injury.

RESULTS: Our deep learning drug screens identified saxagliptin, a dipeptidyl peptidase-4 inhibitor, as a promising awakening drug for both acute and prolonged DOC. The retrospective clinical analysis showed that saxagliptin was associated with the highest recovery rate from acute coma among diabetes medications. After matching patients by age, sex, initial GCS score, coma etiology, and glycemic status, brain-injured patients with diabetes on incretin-based therapies, including dipeptidyl peptidase-4 inhibitors and glucagon-like peptide-1 analogues, recovered from coma at significantly higher rates compared to both brain-injured patients with diabetes on non-incretin-based diabetes medications (95% confidence interval of 1.8-14.1% higher recovery rate, P = 0.0331) and brain-injured patients without diabetes (95% confidence interval of 2-21% higher recovery rate, P = 0.0272). Post matching, brain-injured patients with diabetes on incretin-based therapies also recovered at a significantly higher rate than patients treated with amantadine (95% confidence interval for the difference 2.4-25.1.0%, P = 0.0364). A review of preclinical studies identified several pathways through which saxagliptin and other incretin-based medications may aid awakening from both acute and chronic DOC: restoring monoaminergic and GABAergic neurotransmission, reducing brain inflammation and oxidative damage, clearing hyperphosphorylated tau and amyloid-β, normalizing thalamocortical glucose metabolism, increasing neural plasticity, and mitigating excitotoxic brain damage.

CONCLUSIONS: Our findings suggest incretin-based medications in general, and saxagliptin in particular, as potential novel therapeutic agents for DOC. Further prospective clinical trials are needed to confirm their efficacy and safety in DOC.

PMID:39904872 | DOI:10.1007/s12028-025-02217-0

Genes Cells. 2025 Mar;30(2):e70000. doi: 10.1111/gtc.70000.

ABSTRACT

Cis-regulatory elements (cREs) play a crucial role in regulating gene expression and determining cell differentiation and state transitions. To capture the heterogeneous transitions of cell states associated with these processes, detecting cRE activity at the single-cell level is essential. However, current analytical methods can only capture the average behavior of cREs in cell populations, thereby obscuring cell-specific variations. To address this limitation, we proposed an attention-based deep neural network framework that integrates DNA sequences, genomic distances, and single-cell multi-omics data to detect cREs and their activities in individual cells. Our model shows higher accuracy in identifying cREs within single-cell multi-omics data from healthy human peripheral blood mononuclear cells than other existing methods. Furthermore, it clusters cells more precisely based on predicted cRE activities, enabling a finer differentiation of cell states. When applied to publicly available single-cell data from patients with glioma, the model successfully identified tumor-specific SOX2 activity. Additionally, it revealed the heterogeneous activation of the ZEB1 transcription factor, a regulator of epithelial-to-mesenchymal transition-related genes, which conventional methods struggle to detect. Overall, our model is a powerful tool for detecting cRE regulation at the single-cell level, which may contribute to revealing drug resistance mechanisms in cell sub-populations.

PMID:39904740 | DOI:10.1111/gtc.70000

Med Phys. 2025 Feb 4. doi: 10.1002/mp.17668. Online ahead of print.

ABSTRACT

BACKGROUND: Although deep learning (DL) methods for reconstructing 3D magnetic resonance (MR) volumes from 2D MR images yield promising results, they require large amounts of training data to perform effectively. To overcome this challenge, fine-tuning-a transfer learning technique particularly effective for small datasets-presents a robust solution for developing personalized DL models.

PURPOSE: A 2D to 3D conditional generative adversarial network (GAN) model with a patient- and fraction-specific fine-tuning workflow was developed to reconstruct synthetic 3D MR volumes using orthogonal 2D MR images for online dose adaptation.

METHODS: A total of 2473 3D MR volumes were collected from 43 patients. The training and test datasets were separated into 34 and 9 patients, respectively. All patients underwent MR-guided adaptive radiotherapy using the same imaging protocol. The population data contained 2047 3D MR volumes from the training dataset. Population data were used to train the population-based GAN model. For each fraction of the remaining patients, the population model was fine-tuned with the 3D MR volumes acquired before beam irradiation of the fraction, named the fine-tuned model. The performance of the fine-tuned model was tested using the 3D MR volume acquired immediately after the beam delivery of the fraction. The model's input was a pair of axial and sagittal MR images at the isocenter level, and the output was a 3D MR volume. Model performance was evaluated using the structural similarity index measure (SSIM), peak signal-to-noise ratio (PSNR), root mean square error (RMSE), and mean absolute error (MAE). Moreover, the prostate, bladder, and rectum in the predicted MR images were manually segmented. To assess geometric accuracy, the 2D Dice Similarity Coefficient (DSC) and 2D Hausdorff Distance (HD) were calculated.

RESULTS: A total of 84 3D MR volumes were included in the performance testing. The mean ± standard deviation (SD) of SSIM, PSNR, RMSE, and MAE were 0.64 ± 0.10, 93.9 ± 1.5 dB, 0.050 ± 0.009, and 0.036 ± 0.007 for the population model and 0.72 ± 0.09, 96.2 ± 1.8 dB, 0.041 ± 0.007, and 0.028 ± 0.006 for the fine-tuned model, respectively. The image quality of the fine-tuned model was significantly better than that of the population model (p < 0.05). The mean ± SD of DSC and HD of the population model were 0.79 ± 0.08 and 1.70 ± 2.35 mm for prostate, 0.81 ± 0.10 and 2.75 ± 1.53 mm for bladder, and 0.72 ± 0.08 and 1.93 ± 0.59 mm for rectum. Contrarily, the mean ± SD of DSC and HD of the fine-tuned model were 0.83 ± 0.06 and 1.29 ± 0.77 mm for prostate, 0.85 ± 0.07 and 2.16 ± 1.09 mm for bladder, and 0.77 ± 0.08 and 1.57 ± 0.52 mm for rectum. The geometric accuracy of the fine-tuned model was significantly improved than that of the population model (p < 0.05).

CONCLUSION: By employing a patient- and fraction-specific fine-tuning approach, the GAN model demonstrated promising accuracy despite limited data availability.

PMID:39904621 | DOI:10.1002/mp.17668

Med Phys. 2025 Feb 4. doi: 10.1002/mp.17663. Online ahead of print.

ABSTRACT

BACKGROUND: Kidney ultrasound (US) image segmentation is one of the key steps in computer-aided diagnosis and treatment planning of kidney diseases. Recently, deep learning (DL) technology has demonstrated promising prospects in automatic kidney US segmentation. However, due to the poor quality, particularly the weak boundaries in kidney US imaging, obtaining accurate annotations for DL-based segmentation methods remain a challenging and time-consuming task. This issue can hinder the application of data-hungry deep learning methods.

PURPOSE: In this paper, we explore a novel cross-modal data augmentation method aimed at enhancing the performance of DL-based segmentation networks on the limited labeled kidney US dataset.

METHODS: In particular, we adopt a novel method based on contrastive unpaired translation network (CUT) to obtain simulated labeled kidney US images at a low cost from labeled abdomen computed tomography (CT) data and unlabeled kidney US images. To effectively improve the segmentation network performance, we propose an instance-weighting training strategy that simultaneously captures useful information from both the simulated and real labeled kidney US images. We trained our generative networks on a dataset comprising 4418 labeled CT slices and 4594 unlabeled US images. For segmentation network, we used a dataset consisting of 4594 simulated and 100 real kidney US images for training, 20 images for validation, and 169 real images for testing. We compared the performance of our method to several state-of-the-art approaches using the Wilcoxon signed-rank test, and applied the Bonferroni method for multiple comparison correction.

RESULTS: The experimental results show that we can synthesize accurate labeled kidney US images with a Fréchet inception distance of 52.52. Moreover, the proposed method achieves a segmentation accuracy of 0.9360 ± 0.0398 for U-Net on normal kidney US images, and 0.7719 ± 0.2449 on the abnormal dataset, as measured by the dice similarity coefficient. When compared to other training strategies, the proposed method demonstrated statistically significant superiority, with all p-values being less than 0.01.

CONCLUSIONS: The proposed method can effectively improve the accuracy and generalization ability of kidney US image segmentation models with limited annotated training data.

PMID:39904615 | DOI:10.1002/mp.17663