Category: Statistics

Comput Struct Biotechnol J. 2025 Jul 23;27:3264-3274. doi: 10.1016/j.csbj.2025.07.019. eCollection 2025.

ABSTRACT

Single-cell RNA sequencing (scRNA-seq) enables high-resolution analysis of cellular heterogeneity, yet accurate and consistent cell-type annotation remains a crucial challenge. Numerous automated tools exist, but their complex modeling assumptions can hinder reliability across varied datasets and protocols. We propose PCLDA, a pipeline composed of three modules: t-test-based gene screening, principal component analysis (PCA) and linear discriminant analysis (LDA), all built on simple statistical methods. An ablation study shows that each module in PCLDA contributes significantly to performance and robustness, with two novel enhancements in the second module yielding substantial gains. Despite these additions, the model retains its original assumptions, computational efficiency, and interpretability. Benchmarking against nine state-of-the-art methods across 22 public scRNA-seq datasets and 35 distinct evaluation scenarios, PCLDA consistently achieves top-tier accuracy under both intra-dataset (cross-validation) and inter-dataset (cross-platform) conditions. Notably, when reference and query data are generated via different protocols, PCLDA remains stable and often outperforms more complex machine-learning approaches. Furthermore, PCLDA offers strong interpretability, attributed to the linear nature of its PCA and LDA modules. The final decision boundaries are linear combinations of the original gene expression values, directly reflecting the contribution of each gene to the classification. Top-weighted genes identified by PCLDA better capture biologically meaningful signals in enrichment analyses than those selected via marginal screening alone, offering deeper functional insights into cell-type specificity. In conclusion, our work underscores the utility of carefully enhanced simple statistics methods for single-cell annotation. PCLDA’s simplicity, interpretability, and consistently high performance make it a practical, reliable alternative to more complex annotation pipelines. Code is available on GitHub:https://github.com/kellen8hao/PCLDA.

PMID:40778314 | PMC:PMC12329077 | DOI:10.1016/j.csbj.2025.07.019

Front Aging Neurosci. 2025 Jul 24;17:1617611. doi: 10.3389/fnagi.2025.1617611. eCollection 2025.

ABSTRACT

BACKGROUND: Alzheimer’s Disease (AD) is heterogeneous and shows complex interconnected pathways at various biological levels. Risk scores contribute greatly to disease prognosis and biomarker discovery but typically represent generic risk factors. However, large-scale multi-omics data can generate individualized risk factors. Filtering these risk factors with brain-derived extracellular vesicles (EVs) could yield key pathologic pathways and vesicular vehicles for treatment delivery.

METHODS: A list of 460 EV-related genes was curated from brain tissue samples in the ExoCarta database. This list was used to select genes from transcriptomics, proteomics, and DNA methylation data. Significant risk factors included demographic features (age, sex) and genes significant for progression in transcriptomics data. These genes were selected using Cox regression, aided by the Least Absolute Shrinkage and Selection Operator (LASSO), and were used to construct three risk models at different omics levels. Gene signatures from the significant risk factors were used as biomarkers for further evaluation, including gene set enrichment analysis (GSEA) and drug perturbation analysis.

RESULTS: Nine EV-related genes were identified as significant risk factors. All three risk models predicted high/low risk groups with significant separation in Kaplan-Meier analysis. Training the transcriptomics risk models on EV-related genes yielded better AD classification results than using all genes in an independent dataset. GSEA revealed Mitophagy and several other significant pathways related to AD. Four drugs showed therapeutic potential to target the identified risk factors based on Connectivity Map analysis.

CONCLUSION: The proposed risk score model demonstrates a novel approach to AD using EV-related large-scale multi-omics data. Potential biomarkers and pathways related to AD were identified for further investigation. Drug candidates were identified for further evaluation in biological experiments, potentially transported to targeted tissues via bioengineered EVs.

PMID:40778305 | PMC:PMC12328399 | DOI:10.3389/fnagi.2025.1617611

Front Cell Infect Microbiol. 2025 Jul 24;15:1610445. doi: 10.3389/fcimb.2025.1610445. eCollection 2025.

ABSTRACT

INTRODUCTION: Upper respiratory tract infections (URTIs) represent a significant global health burden, impacting patient morbidity and quality of life. The implementation of accurate pathogen detection methods is crucial for precise diagnosis and effective clinical management of URTIs. This study evaluates the clinical performance of targeted next-generation sequencing (tNGS) as a diagnostic tool for comprehensive identification of URTI-associated pathogens.

METHODS: A total of 190 nasopharyngeal swab specimens from patients were enrolled for the study. These specimens underwent pathogen identification using both tNGS and the FilmArray respiratory panel. The results obtained from these two methods were then compared.

RESULTS: Overall, tNGS identified 164 positive samples, compared to 91 positives identified by FilmArray. Regarding the shared target species or subtypes, tNGS identified 97 positive samples, whereas the FilmArray respiratory panel detected 88 positives out of 190 specimens. tNGS identified a diverse array of 34 different pathogens, significantly surpassing the 12 pathogens identified by the FilmArray panel. The detection rates for tNGS and FilmArray were 51.05% (97/190) and 46.32% (88/190), respectively. Statistical analysis revealed no significant difference in the detection rates of 10 specific respiratory pathogens (with ≥3 positives). Furthermore, the overall pathogen detection accuracy of tNGS was determined to be 90.16% (95%_CI = 83.45%~94.81%), with a sensitivity of 95.45% (88.77%~98.75%).

CONCLUSIONS: The tNGS method demonstrates broader pathogen detection capability compared to the FilmArray, achieving a higher positive detection rate in upper respiratory tract infections. It demonstrates high accuracy and sensitivity, offering a viable and rapid diagnostic approach for upper respiratory tract infections.

PMID:40778290 | PMC:PMC12329660 | DOI:10.3389/fcimb.2025.1610445

Lancet Reg Health Southeast Asia. 2025 Jul 28;40:100643. doi: 10.1016/j.lansea.2025.100643. eCollection 2025 Sep.

ABSTRACT

BACKGROUND: Breast cancer in young adult women is a rapidly growing group of cancer patients in India which needs to be addressed with urgency. Despite increasing global focus on breast cancer in young women, data from India remain scarce. Considering this gap, we undertook this study to analyse the clinicopathological characteristics and prognostic outcomes of young adult women (aged 18-30 years) with breast cancer in Ahmedabad, India.

METHODS: This was a retrospective observational study of a prospectively maintained database of 201 patients with breast cancer (aged 18-30 years) treated in a high-volume tertiary centre in Ahmedabad, India, from January 2015 to December 2020. Patients were followed up until June 2023. The demographic parameters, clinicopathological characteristics and survival of all patients were studied. Statistical analyses were done using SPSS and DATAtab.

FINDINGS: In this study 49.2% of cases were early breast cancers, 26.8% locally advanced, and 23.8% were metastatic. The proportion of aggressive cancers was higher with 38.8% hormone negative, 39.3% HER2-positive, 26.8% triple-negative and 50.8% grade 3. The median overall survival for all patients was 56 months (95% CI 28-84 months) and the 5-year overall survival was 48% (95% CI 40-56%). The multivariate analysis suggested that clinical stage, grade and luminal A status, significantly affected overall survival. The 5-year overall survival and disease-free survival of patients undergoing surgery were 65% (95% CI 57-74%) and 56% (95% CI 47-65%) respectively.

INTERPRETATION: The 5-year overall survival rate of 48% among young adult women with breast cancer included in this study is poor compared to the 77% observed in high-income countries in the western parts of the world. Adoption of appropriate and aggressive treatment strategies may enhance the outcomes in this age group of women with breast cancer.

FUNDING: None.

PMID:40778288 | PMC:PMC12329285 | DOI:10.1016/j.lansea.2025.100643

One Health. 2025 Jul 25;21:101151. doi: 10.1016/j.onehlt.2025.101151. eCollection 2025 Dec.

ABSTRACT

Knowing Echinococcus multilocularis prevalence in wild canids helps us assess the risk of zoonotic spillover. In this study, we updated the infection statistics among red foxes in northern and northeastern Poland since the last monitoring effort at the district (Polish: powiat) level (2001-2004). We collected faecal samples from 192 red foxes from the districts of Słupsk, Puck, Wejherowo, Kościerzyna, and Kartuzy (Pomorskie Voivodship); 200 individuals from the districts of Bartoszyce, Kętrzyn, Gołdap, Iława, and Węgorzewo (Warmińsko-Mazurskie Voivodship); and 47 individuals from the district of Augustów (Podlaskie Voivodship). Based on nested PCR testing confirmed by Sanger sequencing, we calculated the prevalence with a 95 % confidence interval and compared the results to previous reports. We recorded the highest proportion of infected red foxes in Gołdap District (21/40, 52.5 %, 95 % CI: 37.5 %-67.1 %). The shared prevalence in Puck, Wejherowo, Kościerzyna, and Kartuzy Districts was 10.4 % (19/182, 95 % CI: 6.7 %-15.8 %). One red fox tested positive in Iława District (1/48, 2.1 %, 95 % CI: -0.6 %-11.9 %). We found no statistically significant changes in prevalence in the sampled areas. Our results indicate that E. multilocularis prevalence has remained stable in red fox populations throughout the region over the past 20 years.

PMID:40778283 | PMC:PMC12329506 | DOI:10.1016/j.onehlt.2025.101151

Front Endocrinol (Lausanne). 2025 Jul 24;16:1634875. doi: 10.3389/fendo.2025.1634875. eCollection 2025.

ABSTRACT

OBJECTIVE: This study aimed to evaluate the value of constructing a multimodal deep-learning video model based on 2D ultrasound and contrast-enhanced ultrasound (CEUS) dynamic video for the preoperative prediction of OLNM in papillary thyroid carcinoma (PTC) patients.

METHODS: A retrospective analysis was conducted on 396 cases of clinically lymph node-negative PTC cases with ultrasound images collected between January and September 2023. Five representative deep learning architectures were pre-trained to construct deep learning static image models (DL_image), CEUS dynamic video models (DL_CEUSvideo), and combined models (DL_combined). The area under the receiver operating characteristic curve (AUC) was used to evaluate model performance, with comparisons made using the Delong test. A P-value of less than 0.05 was considered statistically significant.

RESULTS: The DL_CEUSvideo, DL_image, and DL_combined models were successfully developed and demonstrated. The AUC values were 0.826 (95% CI: 0.771-0.881), 0.759 (95% CI: 0.690-0.828), and 0.926 (95% CI: 0.891-0.962) in the training set, and 0.701 (95% CI: 0.589-0.813), 0.624 (95% CI: 0.502-0.745), and 0.734 (95% CI: 0.627-0.842) in the test set. Finally, sensitivity, specificity, and accuracy for the DL_CEUSvideo, DL_image, and DL_combined models were 0.836, 0.671, 0.704; 0.673, 0.716, 0.707; and 0.818, 0.902, 0.886 in the training set, and 0.556, 0.775, 0.724; 0.556, 0.674, 0.647; and 0.704, 0.663, 0.672 in the test set, respectively.

CONCLUSION: These results demonstrated that the multimodal deep learning dynamic video model could preoperatively predict OLNM in PTC patients. The DL_CEUSvideo model outperformed the DL_image model, while the DL_combined model significantly enhanced sensitivity without compromising specificity.

PMID:40778281 | PMC:PMC12329689 | DOI:10.3389/fendo.2025.1634875

Front Endocrinol (Lausanne). 2025 Jul 24;16:1564752. doi: 10.3389/fendo.2025.1564752. eCollection 2025.

ABSTRACT

OBJECTIVE: This study aimed to compare the prognosis of unilateral papillary thyroid carcinoma(PTC) patients with ipsilateral cervical lymph node metastasis(IC-LNM) under the treatment of unilateral lobectomy(uLT) vs total thyroidectomy(TT) in order to find out the optimal surgery for these patients without other clinical risk characteristics.

METHODS: PTC patients at Zhejiang Cancer Hospital between 2012 and 2022 were retrospectively reviewed. Additionally, a propensity score matching(PSM) was performed on patients treated with uLT or TT. Recurrence-free survival(RFS), overall survival(OS), hospitalization costs, postoperative complications, and other clinical characteristics were analyzed between the two groups.

RESULTS: Ultimately, 682 unilateral PTC patients with IC-LNM were available in the study. After PSM with possible prognostic factors(such as gender, age, primary tumor size, multifocality, extrathyroidal invasion, and T-stage), 225 pairs of patients were available. With a median of 81(5-154) months follow-up, 22 patients(9.8%) in the uLT and 12(5.3%) in the TT recurred. There were no significant differences in 5-year RFS and 5-year OS between uLT and TT groups. However, TT group was significantly correlated with higher risk of transient and permanent hypoparathyroidism, higher levothyroxine doses, longer hospital stays, and higher hospitalization costs than uLT group(p<0.05).

CONCLUSIONS: Our study indicated that there were no differences in recurrence and survival between unilateral PTC patients with IC-LNM treated with uLT or TT for the primary tumor. However, uLT group had a lower risk of postoperative complications and a lower hospitalization cost than TT group. Thus, for selected unilateral PTC patients with IC-LNM without other risk features, uLT could be recommended.

PMID:40778278 | PMC:PMC12328177 | DOI:10.3389/fendo.2025.1564752

Front Endocrinol (Lausanne). 2025 Jul 24;16:1602844. doi: 10.3389/fendo.2025.1602844. eCollection 2025.

ABSTRACT

BACKGROUND: Thyroid cancer (TC) has shown a rising prevalence worldwide. While numerous studies have explored the relationship between vitamin D levels and TC risk, their conclusions remain inconsistent.

OBJECTIVE: This meta-analysis aims to evaluate the association between serum vitamin D levels, vitamin D deficiency, and TC based on existing evidence.

METHODS: We systematically searched the Embase, Web of Science, and PubMed databases for human studies investigating the relationship between vitamin D and TC including a control group. A random-effects model with forest plots was employed to calculate the mean difference (MD) in serum vitamin D levels, the odds ratio (OR) for vitamin D deficiency, and the risk difference (RD) between TC cases and controls. Meta-regressions and subgroup analyses were conducted based on the season of serum 25(OH)D sampling, source of controls, timing of measurement, study type, and testing methods of 25(OH)D. A p-value <0.05 was considered statistically significant.

RESULTS: A total of 23 studies were included. The meta-analysis revealed that TC patients had significantly lower serum vitamin D compared to the controls [SMD = -0.38 (95% CI: -0.62 to -0.14)].Additionally, vitamin D deficiency was significantly more prevalent among TC patients (OR = 1.33, 95% CI: 1.02 to 1.73, P < 0.05). The subgroup analyses demonstrated significant differences across most subgroups, except for post-operative measurements. Seasonal variation in 25(OH)D sampling was identified as a key source of heterogeneity.

CONCLUSIONS: The meta-analysis suggests that lower serum vitamin D levels and vitamin D deficiency are significantly associated with an increased risk of TC. However, further studies with standardized protocols for seasonal sampling of vitamin D, source of control, measurement timing, study type, and testing methods of 25(OH)D are needed to clarify this relationship and its underlying mechanisms.

PMID:40778274 | PMC:PMC12328179 | DOI:10.3389/fendo.2025.1602844

Front Endocrinol (Lausanne). 2025 Jul 24;16:1595097. doi: 10.3389/fendo.2025.1595097. eCollection 2025.

ABSTRACT

BACKGROUND: The prevalence of hypertension in children is rising globally, with early-onset high blood pressure linked to future cardiovascular risk. Identifying early risk markers beyond obesity and high salt intake is necessary. Although cost-effective indicators of insulin resistance (IR), such as TyG and METS-IR, have been associated with new-onset hypertension in adults, their links with pediatric hypertension, particularly specific phenotypes of hypertension remain unclear.

METHODS: 12,087 individuals aged 7-17 years from the 2017 China National Nutrition and Health Surveillance of Children and Lactating Women were included. Hypertension was defined as systolic blood pressure (SBP) and/or diastolic blood pressure (DBP) ≥95th percentile for sex, age, and height. Isolated systolic hypertension (ISH), defined as SBP ≥95th and DBP <95th percentile. Isolated diastolic hypertension (IDH), defined as DBP ≥95th and SBP <95th percentile. Systolic-diastolic hypertension (SDH), defined as both SBP and DBP ≥95th percentile. The associations of TyG and METS-IR with hypertension phenotypes were investigated using multivariable logistic regression and restricted cubic spline regression.

RESULTS: TyG and METS-IR were positively associated with hypertension and all its phenotypes after multivariable adjustment. Treated as continuous variables, each 1-unit rise in TyG corresponds to 44%, 47%, and 61% higher chance of ISH, IDH, and SDH, respectively (odds ratio [OR]: 1.44, 95% confidence interval [CI]: 1.31-1.59; OR: 1.47, 95%CI: 1.21-1.79; OR: 1.61, 95%CI: 1.35-1.91); each 1-unit rise in METS-IR corresponds to 10%, 6%, and 12% higher chance of ISH, IDH, and SDH, respectively (OR: 1.10, 95%CI: 1.09-1.12; OR: 1.06, 95%CI: 1.03-1.08; OR: 1.12, 95%CI: 1.10-1.14). Consistent positive associations were observed across different subgroups for ISH and SDH, whereas this association for IDH was not statistically significant in several subgroups (e.g., age ≥12 years, sufficient sleep, daily exercise). TyG and METS-IR exhibited linear dose-response relationships with all hypertension phenotypes (p-nonlinear >0.10).

CONCLUSION: TyG and METS-IR show strong relationships with three kinds of hypertension phenotypes. They are promising markers that may contribute to the primary prevention of hypertension in pediatric populations.

PMID:40778272 | PMC:PMC12328147 | DOI:10.3389/fendo.2025.1595097

Eplasty. 2025 Jun 27;25:e22. eCollection 2025.

ABSTRACT

BACKGROUND: Many amputees are left with chronic localized pain, centralized pain, and phantom limb pain or sensation, often resulting from neuromas in the residual limb. Historically, there is no reliably effective intervention for pain associated with neuroma-related residual or phantom limb pain. Targeted muscle reinnervation (TMR) is a surgical procedure first described in 2002 that involves the transfer of residual nerves from amputated limbs to new muscle targets. TMR has been shown to significantly reduce neuroma pain and facilitate the use of prostheses.

METHODS: A prospective study was conducted of 61 patients who underwent TMR for neuroma treatment or prevention between 2017 and 2022. Primary outcomes included overall, phantom, and residual limb pain recorded using the Visual Analog Scale (VAS), as well as Patient-Reported Outcomes Measurement Information System (PROMIS) forms for Pain Intensity, Quality, Interference, and Behavior. Retrospective data was collected for a propensity-matched cohort of non-TMR amputees to compare pain outcomes.

RESULTS: TMR was performed for 25 upper extremity and 35 lower extremity amputations, and 5 patients underwent TMR on multiple limbs. Significant reductions were observed in overall limb pain (-3.2 points), phantom limb pain (-2.6 points), and residual limb pain (-3.0 points) for the TMR cohort. Mean PROMIS scores for TMR patients were 49.7 for Pain Intensity, 54.0 for Pain Quality, 55.3 for Pain Interference, and 56.1 for Pain Behavior. At the 8.4-month follow-up, 43.8% of TMR patients (vs 84% of controls) remained on neuromodulators, opioids, or both, for pain control.

CONCLUSIONS: TMR improved phantom and residual limb pain in amputees, as evidenced by clinically and statistically significant reductions in pain with reduced need for long-term opioids and/or neuromodulators. These findings support the current understanding of TMR but underscore the need for continued investigation to comprehensively assess the potential of this promising technique in improving the functional outcomes and quality of life in the amputee population.

PMID:40778246 | PMC:PMC12331023