ObjectiveTo investigate the mechanism through which miR‑21 improves chronic renal fibrosis in mice via targeted modulation of the phosphatase and tensin homolog （PTEN）/protein kinase B （AKT）/mammalian target of rapamycin （mTOR） pathway. MethodsThirty‑two chronic kidney disease model mice were randomly divided into four groups （n=8 each group）： model group， miR‑21 overexpression group， miR‑21 inhibition group， and miR‑21 inhibition + MK‑2206 group. Eight healthy mice were included as the control group. The miR‑21 overexpression， miR‑21 inhibition， and miR‑21 inhibition + MK‑2206 groups received tail‑vein injections of lentivirus （50 μL， 1×10⁸ TU per mouse） once weekly for three weeks. The control and model groups were injected with an equal volume of empty vector （LV‑NC）. The miR‑21 inhibition + MK‑2206 group additionally received gavage of the AKT/mTOR pathway inhibitor MK‑2206 （480 mg/kg） once weekly for three weeks. The expressions of miR‑21， 24 h urinary protein， serum creatinine （Scr）， blood urea nitrogen （BUN）， and renal tissue levels of collagen Ⅰ， collagen Ⅲ， α‑smooth muscle actin （α‑SMA）， and PTEN protein， as well as p‑AKT/AKT and p‑mTOR/mTOR ratios， were compared among groups. HE staining was used to observe pathological changes in renal tissue， and Masson staining was used to observe the degree of renal fibrosis. A dual‑luciferase assay was performed to verify the targeting relationship between miR‑21 and PTEN. ResultsCompared with the model group， miR‑21 expression in renal tissue increased in the miR‑21 overexpression group （P<0.05） and decreased in the miR‑21 inhibition group （P<0.05）. Compared with the model group， the miR‑21 overexpression group showed increased 24 h urinary protein， Scr， BUN， and renal tissue expression of collagen Ⅰ， collagen Ⅲ， and α‑SMA （all P<0.05）， while these indicators decreased in the miR‑21 inhibition group （P<0.05）. Compared with the miR‑21 inhibition group， the miR‑21 inhibition + MK‑2206 group exhibited lower 24‑h urinary protein， Scr， BUN， and renal tissue expression of Collagen Ⅰ， Collagen Ⅲ， and α‑SMA （all P<0.05）. Compared with the model group， the miR‑21 overexpression group showed decreased PTEN protein expression （P<0.05） and increased p‑AKT/AKT and p‑mTOR/mTOR ratios （P<0.05）， while the miR‑21 inhibition group showed increased PTEN expression （P<0.05） and decreased p‑AKT/AKT and p‑mTOR/mTOR ratios （P<0.05）. Compared with the miR‑21 inhibition group， the miR‑21 inhibition + MK‑2206 group had lower p‑AKT/AKT and p‑mTOR/mTOR ratios （P<0.05）， with no significant difference in PTEN protein expression. HE and Masson staining showed normal kidney structure and almost no fibrosis in the control group. The model group exhibited glomerular enlargement， capillary loop adhesion， and focal fibrosis. The miR-21 overexpression group showed severe destruction of glomerular structure， accompanied by extensive fibrosis and renal tubular atrophy. The pathological changes and degree of fibrosis were alleviated in the miR-21 inhibition group. The miR-21 inhibition + MK-2206 group showed only mild pathological changes and mild fibrosis， with the interstitium being largely normal. Compared with PTEN-WT + NC mimics 1， the relative luciferase activity in the PTEN-WT + miR-21 mimics group decreased （P<0.001）. There was no statistically significant difference in relative luciferase activity between PTEN-WT + NC mimics group and PTEN-MUT + miR-21 mimics group. ConclusionmiR‑21 may improve renal function indicators and alleviate renal fibrosis in chronic kidney disease mice via targeted modulation of PTEN and subsequently inhibiting the AKT/mTOR pathway.

Objective: To explore the value of contrast-enhanced computed tomography (CT) radiomics combined with machine learning algorithms in the preoperative prediction of perineural invasion (PNI) in oral squamous cell carcinoma (OSCC), aiming to provide evidence for assisting clinical treatment decision-making. Methods​: This study was approved by the Ethics Committee of the Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine. A total of 250 OSCC patients confirmed by postoperative pathology were included, comprising 128 PNI-positive and 122 PNI-negative cases. The dataset was randomly divided into training (n=175), validation (n=38), and independent testing (n=37) sets in a ratio of 7:1.5:1.5. Regions of interest were delineated on preoperative images, and radiomic features were extracted. After dimensionality reduction and feature selection using methods like Least Absolute Shrinkage and Selection Operator (LASSO) regression, multiple machine learning models, including support vector machine (SVM), random forest, Light gradient boosting machine (LightGBM), and a Stacking ensemble model, were constructed. Model performance was evaluated using metrics such as the area under the receiver operating characteristic curve (AUC), sensitivity, calibration curves, and decision curve analysis (DCA). Model interpretability was analyzed using Shapley additive explanations (SHAP) and grouped permutation feature importance analysis. Results ​: ​ Among the 250 samples analyzed, the LightGBM model based on radiomics demonstrated the best performance on the independent test set, with an AUC of 0.781, outperforming models like SVM (AUC = 0.730) and Random Forest (AUC = 0.691), as well as clinical models (AUCs ranging 0.549-0.711). The LightGBM model showed good calibration (Brier score 0.198), and DCA indicated high clinical net benefit over a wide threshold probability range. Paired DeLong tests revealed no statistically significant differences in AUC between the ensemble (Stacking) model and the corresponding best-performing radiomics-based model. SHAP analysis and grouped permutation feature importance analysis further indicated that the primary discriminative information for the model came from radiomic texture features. Conclusion ​ The LightGBM model based on contrast-enhanced CT radiomics demonstrated good discriminative ability for preoperative prediction of PNI in OSCC. In the independent test set, it achieved the highest AUC. This model holds promise as a non-invasive auxiliary tool for preoperative risk assessment. Given the limited sample size of the independent test set, these results require further validation in larger cohorts and external datasets.

ObjectiveBased on functional near-infrared spectroscopy (fNIRS), we investigated the brain activity characteristics of gross motor tasks in children with autism spectrum disorder (ASD) and motor dysfunctions (MDs) to provide a theoretical basis for further understanding the mechanism of MDs in children with ASD and designing targeted intervention programs from a central perspective. MethodsAccording to the inclusion and exclusion criteria, 48 children with ASD accompanied by MDs were recruited into the ASD group and 40 children with typically developing (TD) into the TD group. The fNIRS device was used to collect the information of blood oxygen changes in the cortical motor-related brain regions during single-handed bag throwing and tiptoe walking, and the differences in brain activation and functional connectivity between the two groups of children were analyzed from the perspective of brain activation and functional connectivity. ResultsCompared to the TD group, in the object manipulative motor task (one-handed bag throwing), the ASD group showed significantly reduced activation in both left sensorimotor cortex (SMC) and right secondary visual cortex (V2) (P<0.05), whereas the right pre-motor and supplementary motor cortex (PMC&SMA) had significantly higher activation (P<0.01) and showed bilateral brain region activity; in terms of brain functional integration, there was a significant decrease in the strength of brain functional connectivity (P<0.05) and was mainly associated with dorsolateral prefrontal cortex (DLPFC) and V2. In the body stability motor task (tiptoe walking), the ASD group had significantly higher activation in motor-related brain regions such as the DLPFC, SMC, and PMC&SMA (P<0.05) and showed bilateral brain region activity; in terms of brain functional integration, the ASD group had lower strength of brain functional connectivity (P<0.05) and was mainly associated with PMC&SMA and V2. ConclusionChildren with ASD exhibit abnormal brain functional activity characteristics specific to different gross motor tasks in object manipulative and body stability, reflecting insufficient or excessive compensatory activation of local brain regions and impaired cross-regions integration, which may be a potential reason for the poorer gross motor performance of children with ASD, and meanwhile provides data support for further unraveling the mechanisms underlying the occurrence of MDs in the context of ASD and designing targeted intervention programs from a central perspective.

ObjectiveThe proteome of biological evidence contains rich genetic information, namely single amino acid polymorphisms (SAPs) in protein sequences. However, due to the lack of efficient and convenient analysis tools, the application of SAP in public security still faces many challenges. This paper aims to meet the application requirements of SAP analysis for forensic biological evidence’s proteome data. MethodsThe software is divided into three modules. First, based on a built-in database of common non-synonymous single nucleotide polymorphisms (nsSNPs) and SAPs in East Asian populations, the software integrates and annotates newly identified exonic nsSNPs as SAPs, thereby constructing a customized SAP protein sequence database. It then utilizes a pre-installed search engine—either pFind or MaxQuant—to perform analysis and output SAP typing results, identifying both reference and variant types, along with their corresponding imputed nsSNPs. Finally, SAPTyper compares the proteome-based typing results with the individual’s exome-derived nsSNP profile and outputs the comparison report. ResultsSAPTyper accepts proteomic DDA mass spectrometry raw data (DDA acquisition mode) and exome sequencing results of nsSNPs as input and outputs the report of SAPs result. The pFind and Maxquant search engines were used to test the proteome data of 2 hair shafts of2 individuals, and both obtained SAP results. It was found that the results of the Maxquant search engine were slightly less than those of pFind. This result shows that SAPTyper can achieve SAP fingding function. Moreover, the pFind search engine was used to test the proteome data of 3 hair shafts from 1 European person and 1 African person in the literature. Among the sites fully matched by the literature method, sites detected by SAPTyper are also included; for semi-matching sites, that is, nsSNPs are heterozygous, both literature method and SAPTyper method had the risk of missing detection for one type of the allele. Comparing the analysis results of SAPTyper with the SAP test results reported in the literature, it was found that some imputed nsSNP sites identified by the literature method but not detected by SAPTyper had a MAF of less than 0.1% in East Asian populations, and therefore they were not included in the common nsSNP database of East Asian populations constructed by this software. Since the database construction of this software is based on the genetic variation information of East Asian populations, it is currently unable to effectively identify representative unique common variation sites in European or African populations, but it can still identify SAP sites shared by these populations and East Asian populations. ConclusionAn automated SAP analysis algorithm was developed for East Asian populations, and the software named SAPTyper was developed. This software provides a convenient and efficient analysis tool for the research and application of forensic proteomic SAP and has important application prospects in individual identification and phenotypic inference based on SAP.

Objective To identify the independent predictors of programmed death-1 (PD-1) inhibitor-related endocrine adverse events and construct a clinically usable risk prediction model. Methods A total of 302 patients with solid tumors treated with PD-1 inhibitors were retrospectively enrolled. According to the presence or absence of endocrine immune-related adverse events (irAEs), the patients were divided into case group and control group. The clinical and laboratory indexes were compared between the two groups. Multivariable logistic regression was used to confirm independent predictors of endocrine irAEs. The nomogram was constructed, while the receiver operating characteristic (ROC) curve was used to test the prediction performance of the model. Results The overall incidence of endocrine irAEs was 21.9% (66/302), and the incidence of hypothyroidism was 19.5% (59/302). The age, PD-1 inhibitors, free thyroxine, thyroid peroxidase antibody (TPOAb), thyroglobulin, amylase, lymphocyte subset CD3 expression were statistically different between the two groups (P＜0.05). Multivariable logistic regression showed that higher expression of lymphocyte subset CD3 was a protective factor to prevent endocrine irAEs occurrence (P＝0.004), while age＜60 years, higher TPOAb and use of pembrolizumab were independent risk factors of endocrine irAEs (P＜0.05). The nomogram model thus constructed, and when the threshold probability of the model exceeded 0.1, its net benefit was higher. ROC curve showed that the AUC of the model to predict endocrine irAEs was 0.760. The prediction result of the model was highly consistent with the actual result. Conclusions The age, type of PD-1 inhibitor, baseline TPOAb level, and baseline CD3 expression can independently predict endocrine irAEs occurrence or not. The nomogram model based on this model has good predictive efficiency, which can provide reference for early identification of high-risk patients and immunotherapy management.