Curr Res Pharmacol Drug Discov. 2025 Jan 9;8:100212. doi: 10.1016/j.crphar.2024.100212. eCollection 2025.

ABSTRACT

Non-small-cell lung cancer (NSCLC) represents a predominant histological subtype of lung cancer, characterized by high incidence and mortality rates. Despite significant advancements in therapeutic strategies and a deeper understanding of targeted therapies in recent years, tumor resistance remains an inevitable challenge, leading to poor prognostic outcomes. Several studies have indicated that sphingosine kinase 1 (SPHK1) plays a regulatory role in epidermal growth factor receptor (EGFR) signaling, and its elevated expression may be associated with resistance to EGFR tyrosine kinase inhibitors (EGFR-TKIs). Furthermore, the catalytic product of SPHK1, sphingosine 1-phosphate (S1P), along with its receptor, sphingosine 1-phosphate receptor 3 (S1PR3), plays a regulatory role in the function of the EGFR. However, the specific effects of the SPHK1/S1P/S1PR3 axis on EGFR in NSCLC, as well as the combined effects of SPHK1/S1P/S1PR3 inhibitors with the EGFR-TKI gefitinib, remain to be elucidated. In the present study, we investigated the correlation between SPHK1 expression levels and the survival rates of NSCLC patients, the relationship between SPHK1 or S1PR3 and EGFR, and the impact of SPHK1 expression on the half-maximal inhibitory concentration (IC50) of gefitinib in NSCLC. In A549 cells, the phosphorylation of EGFR was significantly reduced following SPHK1 knockdown. Utilizing SPHK1/S1P/S1PR3 inhibitors, namely PF543, TY52156, and FTY720, we established that the SPHK1/S1P/S1PR3 axis modulates EGFR activation in NSCLC. Furthermore, these signaling inhibitors enhanced the anti-proliferative efficacy of the EGFR-TKI gefitinib. RNA sequencing analysis revealed substantial alterations in 85 differentially expressed genes in NSCLC cells treated with the combination of FTY720 and gefitinib. These genes were predominantly associated with pathways such as axon guidance, microRNAs in cancer, and the JAK-STAT signaling pathway, among others. Overall, targeting the SPHK1/S1P/S1PR3 signaling pathway represents a promising therapeutic strategy to enhance gefitinib sensitivity in NSCLC.

PMID:39896887 | PMC:PMC11787445 | DOI:10.1016/j.crphar.2024.100212

bioRxiv [Preprint]. 2025 Jan 20:2025.01.16.633337. doi: 10.1101/2025.01.16.633337.

ABSTRACT

The BioModels Repository contains over 1000 manually curated mechanistic models drawn from published literature, most of which are encoded in the Systems Biology Markup Language (SBML). This community-based standard formally specifies each model, but does not describe the computational experimental conditions to run a simulation. Therefore, it can be challenging to reproduce any given figure or result from a publication with an SBML model alone. The Simulation Experiment Description Markup Language (SED-ML) provides a solution: a standard way to specify exactly how to run a specific experiment that corresponds to a specific figure or result. BioModels was established years before SED-ML, and both systems evolved over time, both in content and acceptance. Hence, only about half of the entries in BioModels contained SED-ML files, and these files reflected the version of SED-ML that was available at the time. Additionally, almost all of these SED-ML files had at least one minor mistake that made them invalid. To make these models and their results more reproducible, we report here on our work updating, correcting and providing new SED-ML files for 1055 curated mechanistic models in BioModels. In addition, because SED-ML is implementation-independent, it can be used for verification , demonstrating that results hold across multiple simulation engines. Here, we use a wrapper architecture for interpreting SED-ML, and report verification results across five different ODE-based biosimulation engines. Our work with SED-ML and the BioModels collection aims to improve the utility of these models by making them more reproducible and credible.

AUTHOR SUMMARY: Reproducing computationally-derived scientific results seems like it should be straightforward, but is often elusive. Code is lost, file formats change, and knowledge of what was done is only partially recorded and/or forgotten. Model repositories such as BioModels address this failing in the Systems Biology domain by encoding models in a standard format that can reproduce a figure from the paper from which it was drawn. Here, we delved into the BioModels repository to ensure that every curated model additionally contained instructions on what to do with that model, and then tested those instructions on a variety of simulation platforms. Not only did this improve the BioModels repository itself, but also improved the infrastructure necessary to run these validation comparisons in the future.

AUTHOR CONTRIBUTIONS: LS: Writing, Conceptualization, Data Curation, Investigation, Methodology, Project Administration, Software, Validation. RMS: Reading, Writing, Data Curation, Methodology TN: Reading, Data Curation, Methodology HH: Reading JK: Conceptualization, Data Curation, Investigation, Methodology, Software. BS: Software LD: Software IIM: Reading, Conceptualization, Funding JCS: Software, Methodology EA: Reading, Writing AAP: Software MLB: Reading, Writing JH: Writing, Methodology EM: Reading, Writing DPN: Reading, Writing, Methodology JG: Reading, Writing, Methodology HMS: Reading, Writing, Funding.

PMID:39896466 | PMC:PMC11785046 | DOI:10.1101/2025.01.16.633337

Naunyn Schmiedebergs Arch Pharmacol. 2025 Feb 3. doi: 10.1007/s00210-025-03838-0. Online ahead of print.

ABSTRACT

PURPOSE: In chronic kidney disease (CKD), raised plasma aldosterone levels are strongly associated with adverse cardiorenal outcomes. Current standard of care may improve outcomes; however, elevated aldosterone levels often persist. We report safety results for BI 690517 (vicadrostat), a potent, selective aldosterone synthase inhibitor under investigation for CKD.

METHODS: Four phase 1 studies of BI 690517 conducted in healthy European/Chinese/Japanese men: two single rising dose (SRD) and two multiple rising dose (MRD) studies.

PRIMARY ENDPOINT: proportion of participants with investigator-defined drug-related adverse events (AEs).

RESULTS: Single and multiple doses of BI 690517 ≤ 80 mg (0.7-80 mg [European SRD]; 3-80 mg [Chinese/Japanese SRD and MRD]) were well tolerated. Proportions of participants with drug-related AEs: European SRD, 8.3% (4/48); Chinese/Japanese SRD, 21.4% (12/56); European MRD, 13.9% (10/72); Japanese MRD, 2.8% (1/36). No serious AEs, deaths, or AEs leading to treatment discontinuation were reported; one AE of severe orthostatic hypotension occurred (European SRD). Plasma exposure to BI 690517 increased dose dependently; median time to maximum concentration was 0.50-1.75 h and mean half-life was 4.4-6.3 h. Exposure was slightly higher in Asians versus Europeans and may relate to lower body weight in Asian participants. A standardized high-fat/high-calorie meal reduced the rate, but not extent, of BI 690517 absorption. Plasma aldosterone concentrations decreased markedly 1-2 h after BI 690517 administration; decreases were more pronounced with increasing BI 690517 doses.

CONCLUSION: BI 690517 was well tolerated and demonstrated dose-dependent inhibition of aldosterone synthesis. Larger studies are warranted to confirm these findings.

PMID:39899058 | DOI:10.1007/s00210-025-03838-0

Int J Med Sci. 2025 Jan 13;22(3):732-744. doi: 10.7150/ijms.103270. eCollection 2025.

ABSTRACT

Aminoglycosides and cisplatin drugs are extensively utilized for their high efficacy in treating various conditions in the clinic, however, their ototoxic side effects warrant significant attention. These drugs could penetrate the inner ear via specific channels or transporters, which not only affect the survival of hair cells but also induce the overproduction of reactive oxygen species. Currently, scientific research mainly addresses this issue through the downstream intervention of reactive oxygen species. However, recent studies have revealed that directly reducing the uptake of these drugs by hair cells can effectively avoid initial damage. In particular, the interactions between drugs and hair cells, as well as the specific functions of relevant channels and transporters, can be explored in detail through the use of molecular dynamics simulations. The swift advancement in the field of structural biology has shed light on the structural functions of various channels and transporters closely related to drug absorption, such as electromechanical transduction channels (MET) and organic cation transporter-2, etc., providing theoretical basis and potential targets for novel ear protection strategies. It is, therefore, imperative to investigate the regulatory role of the MET channel in the up-taking of ototoxic drugs, serving as a pivotal point for the development of preventative and therapeutic approaches. This review aims to highlight the mechanism of inhibition of ototoxic substances absorption by auditory hair cells, explore how to develop novel ear protection methods by targeting these channels and transporters, and provide a new perspective and strategy for addressing drug-induced ototoxicity. The approach to protecting hair cells by targeting these channels and transporters not only broadens our understanding of the underlying mechanisms of ototoxicity, but could also spur further research and progress in the field of auditory protection.

PMID:39898250 | PMC:PMC11783074 | DOI:10.7150/ijms.103270

Cureus. 2024 Dec 31;16(12):e76694. doi: 10.7759/cureus.76694. eCollection 2024 Dec.

ABSTRACT

Most herbal medicines contain licorice, which may inhibit 11-beta-hydroxysteroid dehydrogenase type 2 (11βHSD2). When licorice inhibits 11βHSD2, accumulated cortisol binds excessively to the mineralocorticoid receptor (MR) instead of aldosterone, promoting sodium absorption and potassium excretion. This condition has been called pseudoaldosteronism due to its clinical manifestations resembling those of primary aldosteronism, producing excessive aldosterone. Primary aldosteronism and pseudoaldosteronism usually present with increased urine potassium excretion in the face of hypokalemia due to excessive MR activation by aldosterone and cortisol, respectively, the so-called renal potassium loss. Here, we report an atypical case of licorice-induced pseudoaldosteronism, which unexpectedly presented with decreased urine potassium excretion in the presence of severe hypokalemia in an elderly patient. A spot potassium-to-creatinine ratio was decreased to 4.5 mmol/g creatinine. A 24-hour urine collection also revealed decreased urine potassium excretion of 7.5 mmol/day in the presence of severe hypokalemia. These results on urine potassium were atypical for pseudoaldosteronism. Urine sodium excretion was also significantly reduced to 27.5 mmol/day, suggesting low sodium intake due to anorexia. Low sodium intake followed by low sodium delivery to the collecting duct was the crucial cause of decreased urine potassium excretion in pseudoaldosteronism. Clinicians need to understand the side effects of licorice and assess the risks and benefits associated with the use of licorice-containing drugs, especially in patients with risk factors for pseudoaldosteronism, such as advanced age or low body weight. Careful follow-up of blood pressure or serum potassium concentration is required to prevent the development of pseudoaldosteronism.

PMID:39898153 | PMC:PMC11781896 | DOI:10.7759/cureus.76694

Mol Genet Metab Rep. 2025 Jan 15;42:101189. doi: 10.1016/j.ymgmr.2025.101189. eCollection 2025 Mar.

ABSTRACT

Mucopolysaccharidosis (MPS) IVA (Morquio A syndrome) is an autosomal recessive lysosomal storage disorder caused by a mutation affecting the enzyme N-acetylgalactosamine-6-sulfatase (EC 3.1.6.4, GALNS). Enzyme replacement therapy (ERT) has been shown to improve physical performance, quality of life, and respiratory function in patients with MPS IVA; however, owing to the rarity of MPS IVA, data on Korean patient characteristics are limited. This retrospective study reports clinical, radiographic, biochemical, and molecular findings, and analyzes long-term clinical outcomes, from the largest cohort of Korean patients with MPS IVA in a single center. The analysis included 17 patients from 14 families (58.8 % females; median [range] age at diagnosis 5.2 [1.8-33.7] years). The majority of patients (64.7 %) were classified as having a severe phenotype, 23 % had an intermediate phenotype, and 11.8 % had an attenuated phenotype. Skeletal manifestations and radiologic abnormalities at initial diagnosis included gait abnormality (35.3 %), short stature (23.5 %), chest deformity (23.5 %), scoliosis (17.6 %), kyphosis (11.8 %), dysmorphic face (6 %), hip pain (6 %), and leg deformity (6 %). Twelve different GALNS mutations were identified. Patients received ERT for a median (range) 7.4 years (3.0-12.1). Twelve patients reached final adult height, and all patients with the severe/intermediate phenotype had short stature (<3rd percentile). Hemiepiphysiodesis was the most common surgical intervention among patients with the severe/intermediate phenotype. Drug-related adverse events (urticaria, rash, and anaphylaxis) were reported in four patients but were managed with antihistamines or desensitization. At follow-up, patients experienced improvements in functional independence measure score, ejection fraction, and the 6-min walk test compared with the pre-treatment baseline. This study provides real-world evidence for long-term stabilization of functional independence, endurance, and respiratory function among patients with MPS IVA treated with ERT, with no new safety concerns identified.

PMID:39897469 | PMC:PMC11783393 | DOI:10.1016/j.ymgmr.2025.101189

Cureus. 2025 Jan 2;17(1):e76780. doi: 10.7759/cureus.76780. eCollection 2025 Jan.

ABSTRACT

Angioedema involves fluid accumulation into the interstitial spaces of the dermis, subcutaneous tissue, and mucosal surfaces. While usually benign and self-limited, angioedema can lead to laryngeal edema, a life-threatening condition. The most common causes are histamine-mediated allergic reactions. However, angioedema can also be mediated by bradykinin. Bradykinin-mediated angioedema can occur in the setting of hereditary deficiency of C1q esterase and after exposure to several medications. Drug-induced angioedema is most commonly a secondary complication of non-steroidal anti-inflammatory drug (NSAID) or sulfa drug use. All providers must be aware of the potential side effects of the medications they use or prescribe. We present a case of angioedema resulting from the use of technetium-99m sestamibi tracer injection during an adenosine nuclear stress test.

PMID:39897283 | PMC:PMC11787052 | DOI:10.7759/cureus.76780

Lancet Reg Health West Pac. 2025 Jan 18;55:101464. doi: 10.1016/j.lanwpc.2025.101464. eCollection 2025 Feb.

ABSTRACT

BACKGROUND: Teprotumumab significantly improved proptosis and diplopia in patients with active, moderate-to-severe thyroid eye disease (TED) in previous North American and European studies. This is the first evaluation of efficacy and safety of teprotumumab for active, moderate-to-severe TED in Japanese patients.

METHODS: This randomised, double-masked, placebo-controlled trial was conducted in 16 centres in Japan. Main inclusion criteria were as follows: age 20-80 years; Graves' disease, in a euthyroid or mild hypo/hyperthyroid state; clinical activity score (CAS) ≥3; moderate-to-severe TED; ≥3-mm increase in proptosis before TED onset and/or proptosis ≥18 mm at baseline; and TED duration ≤9 months. Patients were randomised (1:1, stratified by smoking status) to either teprotumumab or placebo. Patients received eight intravenous infusions, one every three weeks for 24 weeks. Patients, investigators, site personnel (except formulating pharmacists) were masked. Primary endpoint was proptosis responder rate (percentage of patients with ≥2-mm proptosis reduction from baseline) at week 24 in the intent-to-treat population. Adverse events were assessed in all patients. This trial was registered at Japan Registry for Clinical Trials (jRCT2031210453).

FINDINGS: Fifty-four patients were randomised (teprotumumab, 27; placebo, 27) between February and November 2022. All patients completed the randomised period, although one teprotumumab patient and two placebo patients missed ≥2 doses. At week 24, the proportion of patients with proptosis response was higher in the teprotumumab group (89%, 24/27) compared with the placebo group (11%, 3/27), 95% confidence interval, 61-95; P<0.0001. Study drug-related adverse events (AEs) occurred in 14 patients (52%) in the teprotumumab group and two patients (7%) in the placebo group; hyperglycaemia-related events were reported in six (22%) and one patient (4%), and hearing impairment in four (15%) and one (4%) patient, respectively. Study drug-related serious AEs and deaths were not reported.

INTERPRETATION: Teprotumumab significantly improved proptosis compared with placebo in Japanese patients with active TED. No study drug-related serious AEs were observed.

FUNDING: Horizon Therapeutics plc (now Amgen).

PMID:39896230 | PMC:PMC11787687 | DOI:10.1016/j.lanwpc.2025.101464

J Am Heart Assoc. 2025 Feb 3:e036882. doi: 10.1161/JAHA.124.036882. Online ahead of print.

ABSTRACT

BACKGROUND: Despite the poor outcomes related to the presence of pulmonary hypertension, it often goes undiagnosed in part because of low suspicion and screening tools not being easily accessible such as echocardiography. A new readily available screening tool to identify elevated pulmonary artery systolic pressures is needed to help with the prognosis and timely treatment of underlying causes such as heart failure or pulmonary vascular remodeling. We developed a deep learning-based method that uses phonocardiograms (PCGs) for the detection of elevated pulmonary artery systolic pressure, an indicator of pulmonary hypertension.

METHODS: Approximately 6000 PCG recordings with the corresponding echocardiogram-based estimated pulmonary artery systolic pressure values, as well as ≈169 000 PCG recordings without associated echocardiograms, were used for training a deep convolutional network to detect pulmonary artery systolic pressures ≥40 mm Hg in a semisupervised manner. Each 15-second PCG, recorded using a digital stethoscope, was processed to generate 5-second mel-spectrograms. An additional labeled data set of 196 patients was used for testing. GradCAM++ was used to visualize high importance segments contributing to the network decision.

RESULTS: An average area under the receiver operator characteristic curve of 0.79 was obtained across 5 cross-validation folds. The testing data set gave a sensitivity of 0.71 and a specificity of 0.73, with pulmonic and left subclavicular locations having higher sensitivities. GradCAM++ technique highlighted physiologically meaningful PCG segments in example pulmonary hypertension recordings.

CONCLUSIONS: We demonstrated the feasibility of using digital stethoscopes in conjunction with deep learning algorithms as a low-cost, noninvasive, and easily accessible screening tool for early detection of pulmonary hypertension.

PMID:39895552 | DOI:10.1161/JAHA.124.036882

Tumori. 2025 Feb 2:3008916251314659. doi: 10.1177/03008916251314659. Online ahead of print.

ABSTRACT

PURPOSE: This study aimed to validate a CT-based radiomics signature for predicting Kirsten rat sarcoma (KRAS) mutation status in lung adenocarcinoma (LADC).

MATERIALS AND METHODS: A total of 815 LADC patients were included. Radiomics features were extracted from non-contrast-enhanced CT (NECT) and contrast-enhanced CT (CECT) images using Pyradiomics. CT-based radiomics were combined with clinical features to distinguish KRAS mutation status. Four feature selection methods and four deep learning classifiers were employed. Data was split into 70% training and 30% test sets, with SMOTE addressing imbalance in the training set. Model performance was evaluated using AUC, accuracy, precision, F1 score, and recall.

RESULTS: The analysis revealed that 10.4% of patients showed KRAS mutations. The study extracted 1061 radiomics features and combined them with 17 clinical features. After feature selection, two signatures were constructed using top 10, 20, and 50 features. The best performance was achieved using Multilayer Perceptron with 20 features. CECT, it showed 66% precision, 76% recall, 69% F1-score, 84% accuracy, and AUC of 93.3% and 87.4% for train and test sets, respectively. For NECT, accuracy was 85% and 82%, with AUC of 90.7% and 87.6% for train and test sets, respectively.

CONCLUSIONS: CT-based radiomics signature is a noninvasive method that can predict KRAS mutation status of LADC when mutational profiling is unavailable.

PMID:39894961 | DOI:10.1177/03008916251314659

J Biomol Struct Dyn. 2025 Feb 3:1-14. doi: 10.1080/07391102.2025.2460069. Online ahead of print.

ABSTRACT

Lung squamous cell carcinoma (LUSC) is a type of non-small cell lung cancer that is the most common and deadly type of lung cancer, originating from the cells lining the bronchi. The progression of LUSC is influenced by various factors, such as genetic, viral, environmental and hormonal factors, immune system response, and smoking history. Despite extensive studies aimed at improving patient survival, the role of gender-specific molecular variants in LUSC progression remains unclear. Using a systems biology approach, combining differential gene expression, network analysis, and machine learning, aberrant mRNA and ncRNAs implicated in LUSC have been identified to improve patient survival, stratify patients and develop novel prognostic strategies. Furthermore, a systematic analysis of the prognostic implications and functional annotations of the molecular variants results in the filtering of key protein-coding genes and non-coding RNAs that are involved in tumor progression. We found several common molecular variants in both genders, including 4 mRNA, 4 miRNAs, and 27 lncRNAs. Among the shared lncRNAs, 5 were novel for both genders. These were found to have a poor prognostic performance in patients with lung cancer. The key players are involved in DNA replication, nucleotide excision repair, complement and coagulation cascades, and estrogen signaling pathways. In this study, we report lncRNAs (PVT1, FAM13A-AS1, LINC00461, NAV2-AS5, PRICKLE2-AS1, and VCAN-AS1) that may function as oncogenes or tumor suppressors by regulating the expression of coding genes, such as RAB24, HECW2, LGR4, and FKBP5. These lncRNAs and coding genes may play important roles in LUSC development and progression.

PMID:39895519 | DOI:10.1080/07391102.2025.2460069

Hum Vaccin Immunother. 2025 Dec;21(1):2452681. doi: 10.1080/21645515.2025.2452681. Epub 2025 Feb 2.

ABSTRACT

Chickenpox outbreaks frequently occur in collective settings such as kindergartens and schools, posing a significant threat to children's physical and mental health. This study aimed to evaluate the immunogenicity and safety of the freeze-dried live attenuated varicella vaccine (VarV) developed by Beijing Minhai Biotechnology Co. LTD. in healthy participants aged 1-12 years. In this phase III, single-center, randomized, double-blind, active-controlled trial,1,200 healthy participants randomly assigned in a 1:1 ratio to receive one dose of either the test vaccine or the active control vaccine. Venous blood samples were collected before vaccination and 42 days after vaccination, and the fluorescent antibody to membrane antigen (FAMA) assay was used to detect VZV antibody. Adverse events (AEs) observed within 42 days after vaccination and serious adverse events (SAEs) within six months after vaccination were recorded. The seroconversion rates in the test and control groups were 96.79% and 96.43%, respectively, with a difference of 0.36% (95% CI, -1.76%-2.48%). The geometric mean titers (GMTs) were 61.74 and 58.04, respectively, with a difference of 1.06 (95% CI, 0.92-1.23). The lower limits of the 95% CI for the differences in seroconversion rates and GMT ratios between the two groups were greater than their respective pre-set non-inferiority margins. The overall incidence of AEs (p = .0112) in the test group was significantly lower than that in the control group. The freeze-dried live attenuated VarV developed by Beijing Minhai Biotechnology Co. LTD. demonstrated good immunogenicity and higher safety compared to the active control vaccine in healthy participants aged 1-12 years.

PMID:39895085 | DOI:10.1080/21645515.2025.2452681

Neurosci Lett. 2025 Jan 31:138146. doi: 10.1016/j.neulet.2025.138146. Online ahead of print.

ABSTRACT

EEG signals exhibit spatio-temporal characteristics due to the neural activity dispersion in space over the brain and the dynamic temporal patterns of electrical activity in neurons. This study tries to effectively utilize the spatio-temporal nature of EEG signals for diagnosing encephalopathy using a combination of novel locality preserving feature extraction using Local Binary Patterns (LBP) and a custom fine-tuned Long Short-Term Memory (LSTM) neural network. A carefully curated primary EEG dataset is used to assess the effectiveness of the technique for treatment of encephalopathies. EEG signals of all electrodes are mapped onto a spatial matrix from which the custom feature extraction method isolates spatial features of the signals. These spatial features are further given to the neural network, which learns to combine the spatial information with temporal dynamics summarizing pertinent details from the raw EEG data. Such a unified representation is key to perform reliable disease classification at the output layer of the neural network, leading to a robust classification system, potentially providing improved diagnosis and treatment. The proposed method shows promising potential for enhancing the automated diagnosis of encephalopathy, with a remarkable accuracy rate of 90.5%. To the best of our knowledge, this is the first attempt to compress and represent both spatial and temporal features into a single vector for encephalopathy detection, simplifying visual diagnosis and providing a robust feature for automated predictions. This advancement holds significant promise for ensuring early detection and intervention strategies in the clinical environment, which in turn enhances patient care.

PMID:39894198 | DOI:10.1016/j.neulet.2025.138146

J Pain Symptom Manage. 2025 Jan 31:S0885-3924(25)00037-5. doi: 10.1016/j.jpainsymman.2025.01.019. Online ahead of print.

ABSTRACT

This review examines the application of natural language processing (NLP) techniques in cancer research using electronic health records (EHRs) and clinical notes. It addresses gaps in existing literature by providing a broader perspective than previous studies focused on specific cancer types or applications. A comprehensive literature search in the Scopus database identified 94 relevant studies published between 2019 and 2024. The analysis revealed a growing trend in NLP applications for cancer research, with information extraction (47 studies) and text classification (40 studies) emerging as predominant NLP tasks, followed by named entity recognition (7 studies). Among cancer types, breast, lung, and colorectal cancers were found to be the most studied. A significant shift from rule-based and traditional machine learning approaches to advanced deep learning techniques and transformer-based models was observed. It was found that dataset sizes used in existing studies varied widely, ranging from small, manually annotated datasets to large-scale EHRs. The review highlighted key challenges, including the limited generalizability of proposed solutions and the need for improved integration into clinical workflows. While NLP techniques show significant potential in analyzing EHRs and clinical notes for cancer research, future work should focus on improving model generalizability, enhancing robustness in handling complex clinical language, and expanding applications to understudied cancer types. The integration of NLP tools into palliative medicine and addressing ethical considerations remain crucial for utilizing the full potential of NLP in enhancing cancer diagnosis, treatment, and patient outcomes. This review provides valuable insights into the current state and future directions of NLP applications in cancer research.

PMID:39894080 | DOI:10.1016/j.jpainsymman.2025.01.019

Comput Biol Med. 2025 Feb 1;187:109740. doi: 10.1016/j.compbiomed.2025.109740. Online ahead of print.

ABSTRACT

Recent advancements in deep learning have revolutionized the field of drug discovery, with Transformer-based models emerging as powerful tools for molecular design and property prediction. However, the lack of explainability in such models remains a significant challenge. In this study, we introduce ABIET (Attention-Based Importance Estimation Tool), an explainable Transformer model designed to identify the most critical regions for drug-target interactions - functional groups (FGs) - in biologically active molecules. Functional groups play a pivotal role in determining chemical behavior and biological interactions. Our approach leverages attention weights from Transformer-encoder architectures trained on SMILES representations to assess the relative importance of molecular subregions. By processing attention scores using a specific strategy - considering bidirectional interactions, layer-based extraction, and activation transformations - we effectively distinguish FGs from non-FG atoms. Experimental validation on diverse datasets targeting pharmacological receptors, including VEGFR2, AA2A, GSK3, JNK3, and DRD2, demonstrates the model's robustness and interpretability. Comparative analysis with state-of-the-art gradient-based and perturbation-based methods confirms ABIET's superior performance, with functional groups receiving statistically higher importance scores. This work enhances the transparency of Transformer predictions, providing critical insights for molecular design, structure-activity analysis, and targeted drug development.

PMID:39894011 | DOI:10.1016/j.compbiomed.2025.109740

Water Res. 2025 Jan 23;275:123192. doi: 10.1016/j.watres.2025.123192. Online ahead of print.

ABSTRACT

Quickly grasping the time-consuming water quality indicators (WQIs) such as total nitrogen (TN) and total phosphorus (TP) of influent is an essential prerequisite for wastewater treatment plants (WWTPs) to prompt respond to sudden shock loads. Soft detection methods based on machine learning models, especially deep learning models, perform well in predicting the normal fluctuations of these time-consuming WQIs but hardly predict their sudden fluctuations mainly due to the lack of extreme fluctuation data for model training. This work employs attention mechanisms to aid deep learning models in learning patterns of anomalous water quality. The lack of interpretability has always hindered deep learning models from optimizing for different application scenarios. Therefore, the local and global sensitivity analyses are performed based on the best-performing attention-based deep learning and ordinary machine learning models, respectively, allowing for reliable feature importance quantification with a small computational burden. In the case study, three types of attention-based deep learning models were developed, including attention-based multilayer perceptron (A-MLP), Transformer composed of stacked A-MLP encoder and A-MLP decoder, and feature-temporal attention-based long short-term memory (FTA-LSTM) neural network with encoder-decoder architecture. These developed attention-based deep learning models consistently outperform the corresponding baseline models in predicting the testing set of TN, TP, and chemical oxygen demand (COD) time series and the anomalous values therein, clearly demonstrating the positive effect of the integrated attention mechanism. Among them, the prediction performance of FTA-LSTM outperforms A-MLP and Transformer (2.01-38.48 % higher R2, 0-85.14 % higher F1-score, 0-62.57 % higher F2-score). Predicting anomalous water quality using attention-based deep learning models is a novel attempt that drives the WWTPs' operation towards being safer, cleaner, and more cost-efficient.

PMID:39893907 | DOI:10.1016/j.watres.2025.123192

Biol Pharm Bull. 2025;48(1):86-92. doi: 10.1248/bpb.b24-00684.

ABSTRACT

Vitamin D plays a crucial role in immune system function. Several studies have indicated that genetic variations in the vitamin D receptor (VDR) and vitamin D binding protein (VDBP, encoded by GC gene) increase the risk of developing asthma. However, the effect of these variations on the prognosis and clinical outcomes of asthma remains unclear. This study, involving 152 adult patients with asthma, aimed to assess the influence of VDR and GC polymorphisms on asthma severity and its exacerbation. Gene polymorphisms previously associated with asthma risk were analyzed, and VDR mRNA expression levels were evaluated in peripheral blood mononuclear cells. The AA genotype of the VDR rs2228570 polymorphism was associated with an elevated risk of severe asthma compared to the AG/GG genotype (odds ratio, 3.20; 95% confidence interval [CI], 1.24-8.28). Furthermore, patients with the rs2228570 AA genotype showed an elevated risk of exacerbation during the 1-year follow-up period (hazard ratio, 4.01; 95% CI, 1.75-9.15). VDR mRNA expression was significantly reduced in patients with the AA genotype. Furthermore, the mRNA expression levels of GLCCI1, HDAC2, NR3C1, and NFE2L2, which are associated with steroid response, were reduced in patients with the AA genotype. Our findings indicate that patients with the AA genotype of VDR rs2228570 are more likely to experience severe asthma and exacerbations. This polymorphism has the potential to reduce vitamin D efficacy by altering VDR function and expression, potentially resulting in increased inflammation and reduced steroid responsiveness in patients with asthma.

PMID:39894560 | DOI:10.1248/bpb.b24-00684

Respir Med. 2025 Jan 31:107980. doi: 10.1016/j.rmed.2025.107980. Online ahead of print.

ABSTRACT

BACKGROUND: Patients with cystic fibrosis (CF) use inhaled medicines daily due to respiratory manifestations. However, only 31% of users is inhaling correctly. Digital solutions targeting inhalation could help CF patients improve their technique and thus health outcomes. However, the use of electronic monitoring devices shows a decrease over time. Therefore, the aim of study was to investigate CF patients' preferences for the use of electronic devices on their inhalation technique on a regular basis and reasons behind these preferences.

METHODS: Semistructured interviews were conducted with 11 CF patients from four European countries to understand their disease history and experiences, daily use of inhaler medication, experiences with digital devices to achieve disease control, and expectations of new devices for monitoring inhalation. A conventional content analysis was applied.

RESULTS: CF patients knew their body well due to their lifelong experiences. However, some patients still experienced periods with more symptoms and need for support. Non-app support was preferred. CF patients reported that digital systems should provide high benefits for regular use. Patients differed in their interest in digital systems for inhalation. Such systems were mostly relevant to CF patients starting a new inhaled treatment/inhaler device or during periods in which the disease was out of control.

CONCLUSIONS: CF patients perceived limited value of digital systems to monitor their inhalation and mostly considered them necessary for specific periods. Extensive experience in using inhalers and existing daily routines to manage a high treatment burden appear involved in limited need of such systems.

PMID:39894083 | DOI:10.1016/j.rmed.2025.107980

Nat Commun. 2025 Feb 2;16(1):1275. doi: 10.1038/s41467-025-56543-0.

ABSTRACT

The biological process of RNA translation is fundamental to cellular life and has wide-ranging implications for human disease. Accurate delineation of RNA translation variation represents a significant challenge due to the complexity of the process and technical limitations. Here, we introduce RiboTIE, a transformer model-based approach designed to enhance the analysis of ribosome profiling data. Unlike existing methods, RiboTIE leverages raw ribosome profiling counts directly to robustly detect translated open reading frames (ORFs) with high precision and sensitivity, evaluated on a diverse set of datasets. We demonstrate that RiboTIE successfully recapitulates known findings and provides novel insights into the regulation of RNA translation in both normal brain and medulloblastoma cancer samples. Our results suggest that RiboTIE is a versatile tool that can significantly improve the accuracy and depth of Ribo-Seq data analysis, thereby advancing our understanding of protein synthesis and its implications in disease.

PMID:39894899 | DOI:10.1038/s41467-025-56543-0

NPJ Digit Med. 2025 Feb 2;8(1):79. doi: 10.1038/s41746-025-01469-6.

ABSTRACT

Body composition prediction from 3D optical imagery has previously been studied with linear algorithms. In this study, we present a novel application of deep 3D convolutional graph networks and nonlinear Gaussian process regression for human body shape parameterization and body composition estimation. We trained and tested linear and nonlinear models with ablation studies on a novel ensemble body shape dataset containing 4286 scans. Nonlinear GPR produced up to a 20% reduction in prediction error and up to a 30% increase in precision over linear regression for both sexes in 10 tested body composition variables. Deep shape features produced 6-8% reduction in prediction error over linear PCA features for males only, and a 4-14% reduction in precision error for both sexes. All coefficients of determination (R2) for all predicted variables were above 0.86 and achieved lower estimation RMSEs than all previous work on 10 metrics of body composition.

PMID:39894882 | DOI:10.1038/s41746-025-01469-6