A prognostic four-gene signature and a therapeutic strategy for hepatocellular carcinoma:Construction and analysis of a circRNA-mediated competing endogenous RNA network

Background:Hepatocellular carcinoma(HCC)has a poor long-term prognosis.The competition of circular RNAs(circRNAs)with endogenous RNA is a novel tool for predicting HCC prognosis.Based on the alterations of circRNA regulatory networks,the analysis of gene modules related to HCC is feasible.Methods:Multiple expression datasets and RNA element targeting prediction tools were used to construct a circRNA-microRNA-mRNA network in HCC.Gene function,pathway,and protein interaction analyses were performed for the differentially expressed genes(DEGs)in this regulatory network.In the proteinprotein interaction network,hub genes were identified and subjected to regression analysis,producing an optimized four-gene signature for prognostic risk stratification in HCC patients.Anti-HCC drugs were excavated by assessing the DEGs between the low-and high-risk groups.A circRNA-microRNA-hub gene subnetwork was constructed,in which three hallmark genes,KIF4A,CCNA2,and PBK,were subjected to functional enrichment analysis.Results:A four-gene signature(KIF4A,CCNA2,PBK,and ZWINT)that effectively estimated the overall survival and aided in prognostic risk assessment in the The Cancer Genome Atlas(TCGA)cohort and International Cancer Genome Consortium(ICGC)cohort was developed.CDK inhibitors,PI3K inhibitors,HDAC inhibitors,and EGFR inhibitors were predicted as four potential mechanisms of drug action(MOA)in high-risk HCC patients.Subsequent analysis has revealed that PBK,CCNA2,and KIF4A play a crucial role in regulating the tumor microenvironment by promoting immune cell invasion,regulating microsatellite instability(MSI),and exerting an impact on HCC progression.Conclusions:The present study highlights the role of the circRNA-related regulatory network,identifies a four-gene prognostic signature and biomarkers,and further identifies novel therapy for HCC.

Identification of an immune classifier for predicting the prognosis and therapeutic response in triple-negative breast cancer

Triple-negative breast cancer(TNBC)poses a significant challenge due to the lack of reliable prognostic gene signatures and an understanding of its immune behavior.Methods:We analyzed clinical information and mRNA expression data from 162 TNBC patients in TCGA-BRCA and 320 patients in METABRIC-BRCA.Utilizing weighted gene coexpression network analysis,we pinpointed 34 TNBC immune genes linked to survival.The least absolute shrinkage and selection operator Cox regression method identified key TNBC immune candidates for prognosis prediction.We calculated chemotherapy sensitivity scores using the“pRRophetic”package in R software and assessed immunotherapy response using the Tumor Immune Dysfunction and Exclusion algorithm.Results:In this study,34 survival-related TNBC immune gene expression profiles were identified.A least absolute shrinkage and selection operator-Cox regression model was used and 15 candidates were prioritized,with a concomitant establishment of a robust risk immune classifier.The high-risk TNBC immune groups showed increased sensitivity to therapeutic agents like RO-3306,Tamoxifen,Sunitinib,JNK Inhibitor VIII,XMD11-85h,BX-912,and Tivozanib.An analysis of the Search Tool for Interaction of Chemicals database revealed the associations between the high-risk group and signaling pathways,such as those involving Rap1,Ras,and PI3K-Akt.The low-risk group showed a higher immunotherapy response rate,as observed through the tumor immune dysfunction and exclusion analysis in the TCGA-TNBC and METABRIC-TNBC cohorts.Conclusion:This study provides insights into the immune complexities of TNBC,paving the way for novel diagnostic approaches and precision treatment methods that exploit its immunological intricacies,thus offering hope for improved management and outcomes of this challenging disease.

Based on necroptosis identifying the immunological features and prognostic signatures of lung adenocarcinoma

Background:Lung adenocarcinoma is one of the most common pathological types of lung malignant tumor with high morbidity and mortality.Long non-coding RNAs are gradually recognized to play crucial roles in tumor occurrence and development.Necroptosis is a newly established way for cell programmed death,undertaking essential roles in anti-tumor.Therefore,identifying necroptosis-related l ong non-coding RNAs and based on them to evaluate the signatures of l ung adenocarcinoma is essential for patients’survival prediction and therapy.Methods:We collected data from the public database and performed the least absolute shrinkage to construct a 13-lncRNAs prognostic model.Based on the Consensus Clustering,ESTIMATE,CIRERSORT,and weighted gene co-expression network analysis to identify the immune signatures.Results:This study identified a 13-lncRNAs prognostic model.The model’s prediction accuracy was evaluated by receiver operating characteristic and independent-prognosis analysis;besides,a Gene Expression Omnibus dataset was applied for external validation.Furthermore,we analyzed the immune features of subgroups in multiple dimensions.A consensus clustering analysis based on the 41 genes was implemented to separate lung adenocarcinoma patients into two subgroups.We compared the features of subgroups in multiple dimensions,including survival,immune microenvironment,immune cells infiltration and gene co-expression network analysis.Conclusion:W e established a prognosis necroptosis-related risk model to predict lung adenocarcinoma patients’prognosis and systematically understood the correlation between immune and necroptosis.This study can applicate in clinical to predict the prognosis of lung adenocarcinoma patients and provide new insight into lung adenocarcinoma immune therapy.

Variable Selection via Biased Estimators in the Linear Regression Model

Least Absolute Shrinkage and Selection Operator (LASSO) is used for variable selection as well as for handling the multicollinearity problem simultaneously in the linear regression model. LASSO produces estimates having high variance if the number of predictors is higher than the number of observations and if high multicollinearity exists among the predictor variables. To handle this problem, Elastic Net (ENet) estimator was introduced by combining LASSO and Ridge estimator (RE). The solutions of LASSO and ENet have been obtained using Least Angle Regression (LARS) and LARS-EN algorithms, respectively. In this article, we proposed an alternative algorithm to overcome the issues in LASSO that can be combined LASSO with other exiting biased estimators namely Almost Unbiased Ridge Estimator (AURE), Liu Estimator (LE), Almost Unbiased Liu Estimator (AULE), Principal Component Regression Estimator (PCRE), r-k class estimator and r-d class estimator. Further, we examine the performance of the proposed algorithm using a Monte-Carlo simulation study and real-world examples. The results showed that the LARS-rk and LARS-rd algorithms,?which are combined LASSO with r-k class estimator and r-d class estimator,?outperformed other algorithms under the moderated and severe multicollinearity.

Integrated analysis of single-cell and bulk RNA-seq establishes a novel signature for prediction in gastric cancer

BACKGROUND Single-cell sequencing technology provides the capability to analyze changes in specific cell types during the progression of disease.However,previous single-cell sequencing studies on gastric cancer(GC)have largely focused on immune cells and stromal cells,and further elucidation is required regarding the alterations that occur in gastric epithelial cells during the development of GC.AIM To create a GC prediction model based on single-cell and bulk RNA sequencing(bulk RNA-seq)data.METHODS In this study,we conducted a comprehensive analysis by integrating three singlecell RNA sequencing(scRNA-seq)datasets and ten bulk RNA-seq datasets.Our analysis mainly focused on determining cell proportions and identifying differentially expressed genes(DEGs).Specifically,we performed differential expression analysis among epithelial cells in GC tissues and normal gastric tissues(NAGs)and utilized both single-cell and bulk RNA-seq data to establish a prediction model for GC.We further validated the accuracy of the GC prediction model in bulk RNA-seq data.We also used Kaplan–Meier plots to verify the correlation between genes in the prediction model and the prognosis of GC.RESULTS By analyzing scRNA-seq data from a total of 70707 cells from GC tissue,NAG,and chronic gastric tissue,10 cell types were identified,and DEGs in GC and normal epithelial cells were screened.After determining the DEGs in GC and normal gastric samples identified by bulk RNA-seq data,a GC predictive classifier was constructed using the Least absolute shrinkage and selection operator(LASSO)and random forest methods.The LASSO classifier showed good performance in both validation and model verification using The Cancer Genome Atlas and Genotype-Tissue Expression(GTEx)datasets[area under the curve(AUC)_min=0.988,AUC_1se=0.994],and the random forest model also achieved good results with the validation set(AUC=0.92).Genes TIMP1,PLOD3,CKS2,TYMP,TNFRSF10B,CPNE1,GDF15,BCAP31,and CLDN7 were identified to have high importance values in multiple GC predictive models,and KM-PLOTTER analysis showed their relevance to GC prognosis,suggesting their potential for use in GC diagnosis and treatment.CONCLUSION A predictive classifier was established based on the analysis of RNA-seq data,and the genes in it are expected to serve as auxiliary markers in the clinical diagnosis of GC.

纤维肌痛综合征生物标记物的筛选及免疫细胞浸润分析

背景:纤维肌痛综合征作为常见风湿病,其发病与中枢敏化及免疫异常有关,但具体过程尚未阐明,缺乏特异性诊断标志物,不断探索该病的发病机制具有重要的临床意义。目的:基于加权基因共表达网络分析(WGCNA)等生物信息学方法和机器学习算法筛选纤维肌痛综合征潜在的诊断相关标志基因,并分析其免疫细胞浸润特征。方法:对来自基因表达综合数据库(GEO)的纤维肌痛综合征数据集转录谱进行差异分析和WGCNA分析,整合筛选出差异共表达基因,进一步采用机器学习套索回归(LASSO)算法、支持向量机递归特征消除(SVM-RFE)机器学习算法来识别核心生物标志物,并绘制受试者工作特征(ROC)曲线以评估诊断价值。最后,采用单样本基因集富集分析(ssGSEA)和基因集富集分析(GSEA)评估纤维肌痛综合征的免疫细胞浸润情况及通路富集。结果与结论:①对GSE67311数据集按照log2|(FC)|>0,P<0.05的条件进行差异分析后获得8个下调的差异表达基因;进行WGCNA分析后获得正相关性最高(r=0.22,P=0.04)的模块(MEdarkviolet)内含基因497个,负相关性最高(r=-0.41,P=6×10-5)的模块(MEsalmon2)内含基因19个;将差异表达基因与WGCNA的2个高相关性模块基因取交集,获得7个基因。②对上述7个基因进行LASSO回归算法筛选出4个基因,进行SVM-RFE机器学习算法筛选出5个基因,两者取交集后确定了3个核心基因,分别为重组1号染色体开放阅读框150蛋白(germinal center associated signaling and motility like,GCSAML)、整合素β8(Integrin beta-8,ITGB8)和羧肽酶A3(carboxypeptidase A3,CPA3);绘制3个核心基因的ROC曲线下面积分别为0.744,0.739,0.734,提示均具有很好的诊断价值,可作为纤维肌痛综合征的生物标志物。③免疫浸润分析结果显示,与对照组相比纤维肌痛综合征患者记忆B细胞、CD56 bright NK细胞和肥大细胞显著下调(P<0.05),且与上述3个生物标志物显著正相关(P<0.05)。④富集分析结果提示,纤维肌痛综合征的富集途径包括9条,主要与嗅觉传导、神经活性配体-受体相互作用及感染等通路密切相关。⑤上述结果显示,纤维肌痛综合征的发生发展与多基因参与、免疫调节异常及多个通路失调有关,但这些基因与免疫细胞之间的相互作用,以及它们与各通路之间的关系尚需进一步研究。

基于多变量模式分析的飞行学员脑功能连接的识别研究

目的基于多变量模式分析(multivariate pattern analysis,MVPA)对飞行学员和健康的普通人的大脑功能连接进行有效识别。材料与方法采集了40名已经取得执照的飞行专业在校学生与39名地面专业在校学生的功能磁共振数据。通过网络功能连接分析得到功能连接矩阵作为特征,分别通过最小绝对收缩选择算子(least absolute shrinkage and selection operator,LASSO)算法与独立样本t检验方法对特征降维。使用不同核函数的支持向量机(support vector machine,SVM)进行训练和预测,使用留一交叉验证法进行模型性能评估,最终根据训练后SVM模型中的权重定位对应脑区之间的功能连接。结果使用LASSO特征筛选的线性(linear)核SVM模型准确率为81.82%,敏感度82.05%,特异度81.58%,曲线下面积(area under the curve,AUC)为0.88。核函数对模型准确率的影响不大。模型中右侧中央旁小叶、双侧中央后回、双侧顶下缘角回、右侧梭状回、左侧眶部额中回、左侧顶上回、右侧眶部额下回有较高的权重,模型中的权重集中在感觉运动网络(somatomotor network,SMN)与默认模式网络(default mode network,DMN),分别占用所有权重的25.62%和25.27%。结论结合LASSO算法进行特征筛选的SVM可以对飞行学员大脑进行有效识别,并且有更好的可解释性和更小的过拟合。模型权重信息反映了飞行学员主要在运动能力和感知能力有别于普通人。

基于LASSO回归的宁夏回族自治区不同学段儿童青少年近视影响因素分析

目的分析宁夏回族自治区儿童青少年近视流行现状、影响因素及不同学段间的差异。方法采用分层整群随机抽样的方法,于2019年9月至12月,在宁夏回族自治区银川市、吴忠市、石嘴山市、固原市和中卫市,随机抽取8所小学、6所初中、6所高中、4所大学的学生为研究对象,小学每个年级抽取5个班级,初中至大学每个年级抽取4个班级,以抽取班级的全体学生作为研究对象,共抽取学生14211人,对其进行问卷调查、体格检查和视力测量。不同学段儿童近视的影响因素采用最小绝对收缩和选择算子(LASSO)联合Logistic回归进行分析,选择贝叶斯信息准则(Bayesian information criterion,BIC)最小的模型为最优模型。结果宁夏回族自治区儿童青少年近视检出率为70.3%,女生高于男生,城市高于乡镇,差异均有统计学意义(均为P<0.001);按学段分层后,随着年级的增加,近视检出率随之升高,小学最低,大学最高,不同学段近视检出率差异有统计学意义(P<0.001)。近视影响因素的LASSO-Logistic回归分析表明,城乡、性别、年龄、目前是否配戴眼镜、每日课间操节数、是否积极参加体力活动和过去6个月是否保持规律活动是小学生近视的影响因素(均为P<0.05);性别、目前是否配戴眼镜是初中生和高中生近视的影响因素(均为P<0.05);目前是否配戴眼镜是大学生近视的影响因素(P<0.05)。结论宁夏回族自治区儿童青少年近视检出率高,不同学段儿童青少年近视影响因素差异明显。配戴眼镜是控制近视的保护因素。应根据儿童青少年所处学段开展有针对性的视力相关知识的健康教育,增强其健康保健意识,提高儿童青少年视力。

基于X线的纹理分析在诊断跟距联合畸形中的临床应用价值

目的构建跟距联合畸形(talocalcaneal coalition)的X线影像组学模型,并检验其对跟距联合畸形的筛查诊断能力。方法回顾性分析2019年1月至2023年3月吉林大学中日联谊医院放射线科200例行踝关节或足部X线检查的患者临床放射资料(跟距联合阳性及阴性各100例),手动勾画跟距联合畸形所在影像学区域,基于Python-pyradiomics库初步提取影像组学特征,通过曼-惠特尼U检验及最小绝对收缩和选择算子(least absolute shrinkage and selection operator,LASSO)算法实现数据降维和特征筛选,用支持向量机(support vector machine,SVM)对筛选得到的影像组学特征分类建模,最终以受试者工作特征(receiver operating characteristic,ROC)曲线的曲线下面积(area under the curve,AUC)、精确度、召回率、敏感度、特异度及F1分数评价模型的诊断效能。结果从X线图像中初步提取到105个组学特征,经曼-惠特尼U检验及LASSO算法筛选出7个强相关性特征,最终以SVM分类器所构建模型的测试集AUC值为0.93,精确度、召回率、敏感度、特异度和F1分数分别为88%、85%、93%、92%、88%,对跟距联合畸形有良好的筛查诊断能力。结论基于X线的影像组学模型可作为筛查诊断跟距联合畸形的一种准确高效的无创性工具,帮助临床医师诊断跟距联合畸形。

高尿酸血症与慢性肺源性心脏病的相关性研究:基于LASSO回归与倾向性评分匹配法

背景近年来众多研究表明高尿酸血症(HUA)是某些疾病的影响因素,然而HUA是否为慢性肺源性心脏病(CPHD)的影响因素仍需进一步研究。目的探讨HUA与CPHD的相关性,旨在为CPHD患者血尿酸(SUA)水平的管理提供理论依据。方法纳入2019—2023年新疆医科大学第一附属医院收治的1171例慢性阻塞性肺疾病(COPD)患者为研究对象,根据其是否患有CPHD分为CPHD组(470例)和COPD组(701例)。收集患者一般资料和实验室检查及超声心动图检查结果。采用LASSO回归法对变量进行筛选,采用倾向性评分匹配法(PSM)排除混杂因素影响。采用多因素Logistic回归分析探究COPD患者合并CPHD的影响因素。结果CPHD组女性、汉族、吸烟、饮酒、特发性肺纤维化、慢性支气管炎、支气管哮喘比例、淋巴细胞百分比、左心室舒张末期内径、左心室收缩末期内径、心输出量、左心室射血分数低于COPD组,心功能3~4级、HUA、肺栓塞、先天性心脏病比例、红细胞计数、中性粒细胞百分比、SUA、血尿素氮、D-二聚体、N末端-B型利钠肽前体、右心房内径、右心室内径、左心房内径、右心室流出道内径、肺动脉内径高于COPD组,差异有统计学意义(P<0.05)。LASSO回归筛选出变量后进行PSM,最终得到COPD组469例、CPHD组469例。匹配后CPHD组心功能3~4级、HUA占比、右心房内径、右心室内径、右心室流出道内径、肺动脉内径大于COPD组,支气管哮喘、淋巴细胞百分比低于COPD组,差异有统计学意义(P<0.05)。多因素Logistic回归分析结果显示,HUA升高、心功能3~4级、右心房内径、右心室内径、肺动脉内径增加是COPD患者合并CPHD的危险因素(P<0.05),患有支气管哮喘、左心室舒张末期内径增加为COPD患者合并CPHD的保护因素(P<0.05)。将SUA水平按四分位数分层,多因素Logistic回归分析结果显示,与Q1(SUA<237.31μmol/L)比较,Q4(SUA>381.29μmol/L)患者患有CPHD的风险增加1.421倍。结论HUA是CPHD疾病发生、发展的影响因素,积极控制SUA水平有助于预防CPHD的发生、发展。

多传感器信息融合的轴承故障迁移诊断方法

在重型装备低速、重载、强噪声环境下,采用单一传感器难以全面获取轴承的故障诊断信息,导致故障识别率低、识别不稳定,致使变工况下轴承故障迁移诊断失效。针对以上问题,提出了一种多传感器信息融合的轴承故障迁移诊断方法。首先,结合传感器的通道数,构建了堆叠卷积神经网络(MCNNs)提取各个通道的故障特征;然后,在MCNNs中引入最小绝对收缩与选择算子(Lasso),并通过网络反向传播完成了特征权值的更新,从而获得了多通道特征的融合;最后,利用源域数据对模型进行了训练,提取了故障特征,并完成了特征融合,采用损失函数完成了模型参数的优化,将源域训练得到的模型结果作为目标域的初始模型,利用目标域样本对初始模型的参数进行了微调,从而完成了模型迁移;并进行了信息融合效果、方法对比以及传感器信息采集属性的性能实验。研究结果表明:传感器的安装位置对信息融合影响较大,MCNNs+Lasso方法具有较好的特征融合效果,平均迁移诊断精度为99.03%,部分精度可达99.97%,在多个变工况的迁移任务中表现出较高迁移精度和良好的泛化性能。

基于Nomogram模型鉴别肺腺癌病理亚型的临床价值

目的 探讨基于最小绝对收缩和选择算子(least absolute shrinkage and selection operator,LASSO)回归分析构建Nomogram模型预测原位腺癌(AIS)、微浸润腺癌(MIA)及浸润性腺癌(IAC)的价值。方法 选取本院97例经手术病理证实且病理亚型明确的肺腺癌患者,将AIS和MIA归为第1组,IAC为第2组,比较两组患者年龄、性别、吸烟史、长径、短径及免疫组化Ki-67等临床医学特征差异,采用3D Slicer软件进行图像分割,特征提取与选择,通过LASSO算法对特征进行降维,筛选影像组学特征构建预测模型。再采用R软件的rms工具包构建Nomogram模型,计算ROC曲线下面积(AUC),以评价Nomogram模型鉴别肺磨玻璃结节病理亚型的效能。结果 1)性别、吸烟史、长径、短径及免疫组化Ki-67等临床医学特征均差异无统计学意义(P>0.05);2)筛选7个CT影像组学特征:平面度、大依赖低灰度强调、小波变换LHL第十百分位、小波变换HLL第十百分位、小波变换最小值、小波变换均值及小依赖低灰度强度比较,差异均有统计学意义(P均<0.05);3)基于CT影像组学特征建立预测肺磨玻璃结节病理亚型的Nomogram模型,训练集中AUC为0.863,准确率为87.9%,灵敏度为67.9%,特异度为91.1%;验证集中AUC为0.792,准确率为75.0%,灵敏度为66.7%,特异度为90.5%,可见此Nomogram模型具有较好的预测效能。结论 对于预测肺腺癌浸润程度,Nomogram模型具有明显优势,可作为一种鉴别手段。

基于DGA与TPE-LightGBM的变压器故障诊断

油中溶解气体分析(dissolved gas analysis,DGA)对变压器故障的早期预警及诊断具有重要意义。为了提升变压器故障诊断的准确性及可靠性,提出一种基于树结构概率密度估计(tree-structured parzen estimator,TPE)算法优化轻量级梯度提升机(light gradient boosting machine,LightGBM)的变压器故障诊断方法。首先,建立包含油中气体比值、编码等16维DGA特征集合,采用最小绝对收缩和选择(least absolute shrinkage and selection opera-tor,LASSO)算法选择用于变压器故障诊断的有效特征量;其次,构建基于LightGBM的变压器故障诊断方法,并引入TPE算法对LightGBM诊断模型参数进行优化,形成最优故障诊断模型;最后,选用精确度、召回率和F1分数等评价指标对所提诊断模型性能进行评估。研究结果表明,TPE-LightGBM的平均准确率为90.23%,其诊断精度及鲁棒性均优于RF和XGBoost等算法。同时,与现场常用的三比值法进行对比,所提方法的准确性和可靠性均有显著提升。该方法可有效提升电力变压器的智能运维水平。

基于自噬基因的度洛西汀抗抑郁疗效预测模型的构建

目的:通过生物信息学方法构建基于自噬基因的度洛西汀抗抑郁疗效预测模型。方法:在高通量基因表达数据库中下载GSE146446数据集,该芯片包括96例患者接受抗抑郁药物度洛西汀8周的治疗,组织样本为全血样本,以度洛西汀治疗8周后是否有效分组,筛选两组间的差异表达基因,与自噬基因集取交集。利用最小绝对值收敛和选择算法回归(LASSO)及Logistic回归构建疗效预测模型。结果:SPNS1、ITPR3基因的表达水平均为度洛西汀抗抑郁疗效的影响因素(P均<0.05)。LASSO-Logistic回归模型:Logit(P)=33.7846+(-2.8615×SPNS1表达水平)+(-1.7716×ITPR3表达水平),其中Logit(P)=ln[P/(1-P)]。结论:基于自噬相关基因(SPNS1、ITPR3)表达量的度洛西汀的抗抑郁疗效预测模型具有较好的区分度、校准度以及疗效预测效能,未来可能为抑郁症患者使用度洛西汀药物治疗提供更为科学可靠的证据。

基于LASSO回归和Nomogram预测经皮肾活检术后出血风险

目的探讨超声引导下经皮肾穿刺活检术后出血的危险因素,并构建列线图预测模型。方法回顾性收集2022年6月1日至2023年6月1日在十堰市太和医院超声医学科进行实时超声引导下经皮肾穿刺活检住院患者的临床及影像资料。采用LASSO回归、Logistic回归分析,构建超声引导下肾穿刺活检术后出血的列线图预测模型,利用受试者工作特征(ROC)曲线、校正曲线(calibration curve)和决策曲线分析(DCA)三个层面对模型进行评估。结果最终纳入206例超声引导下肾穿刺活检的患者。LASSO回归及Logistic回归分析结果显示,高血压病史(OR=5.339,P<0.001)、穿刺肾下极皮质厚度(OR=0.410,P<0.001)、穿刺肾皮髓质分界不清(OR=6.133,P<0.001)和穿刺时患者不能配合(OR=4.525,P=0.004)是超声引导下肾穿刺活检后出血的独立危险因素。列线图预测模型具有良好的诊断效能AUC=0.891,95%CI为0.842~0.941,绘制模型校准曲线,平均绝对误差为0.026,理想曲线和校正曲线贴合较好。Hosmer-Lemeshow检测χ^(2)=6.599,P=0.580(P>0.05),表明该模型的准确度较好。绘制临床决策曲线显示当列线图预测模型的阈概率小于89%时,该模型的临床净收益率最高。结论高血压病史、穿刺肾下极皮质厚度、穿刺肾皮髓质分界不清、穿刺时患者不能配合是超声引导下肾穿刺活检术后出血的危险因素;基于列线图模型预测超声引导下肾穿刺活检术后出血具有可行性,可以为临床评估肾穿刺活检后出血风险提供可视化依据。

早期帕金森病诊断评分模型构建及效能验证

目的构建早期帕金森病(PD)的诊断评分模型,并验证其效能。方法选择PD患者75例及性别、年龄与PD患者相匹配的健康志愿者75例,随机分为验证组(PD患者38例、健康志愿者37例)与训练组(PD患者37例、健康志愿者38例)。收集受试者病历资料。用最小绝对收缩和选择算子(LASSO)算法,通过十折交叉验证确定最优参数,从训练组相关资料中筛选出具有相关性的诊断因子,并根据各因子系数构建诊断评分模型。通过Logistic回归构建列线图;绘制受试者工作特征曲线,通过曲线下面积和校准曲线评价该模型的诊断效能以及拟合度。结果训练组与验证组相关资料比较差异无统计学意义(P均>0.05)。训练组经LASSO算法确定最佳参数λ=0.052,筛选出具鉴别能力的7个指标,诊断评分模型公式=-1.048+0.961×睡眠行为障碍筛查问卷(RBDSQ)评分+0.079×汉密尔顿焦虑量表14项(HAMA-14)评分-0.0002×神经元特异性烯醇化酶(NSE)-0.011×血管内皮生长因子(VEGF)-0.001×尿酸-0.046×各向异性(FA)+0.003×舒张末期血流速度(DFV)。多因素Logistic回归分析确认所筛选的7个指标可作为早期PD患者的独立诊断因子。在验证组中该诊断评分模型用于诊断早期PD患者的曲线下面积为0.91,高于7个因子单独诊断早期PD的曲线下面积;拟合曲线显示该模型有较好的拟合优度。结论基于RBDSQ评分、HAMA-14评分、VEGF、FA、NSE、尿酸及DFV构建了早期PD的诊断评分模型,该模型有较高的诊断效能。

凝血指标对髋部骨折患者血栓形成风险的预测

目的探讨凝血指标对髋部骨折患者围术期静脉血栓形成风险的预测价值。方法回顾性纳入2020年2月至2022年12月收入笔者医院的160例髋部骨折患者,采取随机数字表法将其分为训练集(n=112例)和验证集(n=48例),进一步根据训练集患者围术期是否发生深静脉血栓(deep vein thrombosis,DVT)划分为发生组和未发生组。使用最小绝对收缩与选择算子(least absolute shrinkage and selection operator,LASSO)选取变量形成LASSO回归模型;绘制受试者操作特征(receiver operating characteristic,ROC)曲线分析模型的预测效能;通过验证集数据实现验证模型的预测效能。结果训练集与验证集患者性别、年龄等一般及临床资料比较,差异均无统计学意义(P>0.05)。训练集内发生组与未发生组患者性别、年龄、骨折至入院时间、体质量指数、D-二聚体、纤维蛋白原、总蛋白、白蛋白、前白蛋白、球蛋白、血红蛋白、血清钙、红细胞体积、白细胞计数、红细胞体积分布宽度、活化部分凝血活酶时间、凝血酶原时间及淋巴细胞比率比较,差异有统计学意义(P<0.05)。通过组间相关系数(intraclass correlation coefficient,ICC)与LASSO筛选出6个非零系数的最优变量,分别是年龄、体质量指数、D-二聚体、纤维蛋白原、凝血酶原时间及活化部分凝血活酶时间,非零系数依次为9.104、1.792、1.270、2.447、3.037及-1.561。结论年龄、体质量指数、D-二聚体、纤维蛋白原、凝血酶原时间及活化部分凝血活酶时间变量联合形成的LASSO回归模型可作为预测髋部骨折患者围术期DVT形成风险的辅助工具。

Characterization of immunogenic cell death-related genes predicting prognosis in colon adenocarcinoma

Background:Colon adenocarcinoma(COAD)is a gastrointestinal malignancy with a high mortality rate.Studies have confirmed the role of immunogenic cell death(ICD)in different cancer types.However,there is a lack of research on ICD-related genes(ICD-RGs)in COAD.This study aimed to examine the impact of ICD-RGs on COAD and their interaction with the immune microenvironment.Methods:Using data from The Cancer Genome Atlas and Gene Expression Omnibus databases,we identified 107 ICD-RGs in COAD.Using a one-way Cox regression analysis,we examined the relationship between these ICD-RGs and overall survival in COAD.Results:Following the regression analyses,we identified 14 overall survival-related genes.Furthermore,we examined the predictive impact of the ICD-RGs using the least absolute shrinkage and selection operator regression analysis and developed a nine-genes prognostic model.The Cancer Genome Atlas and Gene Expression Omnibus datasets were used for training and validation.Kaplan-Meier analysis was used to confirm that the high-risk group had a lower survival rate than the low-risk group.Finally,following a multifactorial analysis,we created a prognostic nomogram that integrated clinical data and risk scores.Conclusions:The nine-genes model exhibits robust stability and can provide valuable insights for guiding the development of tumor immunotherapy strategies and personalized drug selection for patients with COAD.

基于混合级联模型的现货市场日前电价预测方法

准确预测节点电价有助于调度机构实现对发电机组的最优调度,实现电力系统发电成本最优的目的,同时有利于发电方把握市场走向,构建最优的电量、电价投标策略以获取最大利润。由于用电行为、天气情况、电网调度等因素的影响,节点电价在不同时期的数据分布差异较大,预测难度高。因此,提出一种基于混合级联模型的日前节点电价预测方法,采用不同模型分别对突变天和非突变天的节点电价进行预测,实验证明该方法可以有效提高节点电价预测精度。首先,定义突变天,并对历史实际节点电价数据进行标注;其次,获取全省的气象、市场等数据,作为模型的输入特征并进行预处理,根据上述特征构建二分类模型判断预测日是否属于突变天;最后,根据二分类结果,将预测日数据输入最小绝对收缩和选择算法模型或随机森林回归模型,预测节点电价。结果验证了所提方法的有效性和优越性。

Discrimination of Acori Tatarinowii Rhizoma from two habitats based on GC-MS fingerprinting and LASSO-PLS-DA

This study is intended to explore the chemical differences of Acori Tatarinowii Rhizoma （ATR） samples collected from two habitats, Sichuan and Anhui provinces, China. Gas chromatography-mass spectrometry （GC-MS） was applied to establishing the quantitative chemical fingerprints of ATRs. A total of 104 volatile compounds were identified and quantified with the information of mass spectra and retention index （RI）. Furthermore, least absolute shrinkage and selection operator （LASSO）, a sparse regularization method, combined with subsampling was employed to improve the classification ability of partial least squares-discriminant analysis （PLS-DA）. After variable selection by LASSO, three chemical markers,β-elemene, α-selinene and α-asarone, were identified for the discrimination of ATRs from two habitats, and the total classification correct rate was increased from 82.76% to 96.55%. The proposed LASSO-PLS-DA method can serve as an efficient strategy for screening marked chemical components and geo-herbalism research of traditional Chinese medicines.