目的使用不同机器学习算法开发高龄骨科患者术后低蛋白血症风险预测模型。方法使用中国中医科学院望京医院骨科患者数据库作为数据来源;根据低蛋白血症诊断标准,选取数据库中的低蛋白血症患者,提取其人口学特征、生命体征、实验室检查...目的使用不同机器学习算法开发高龄骨科患者术后低蛋白血症风险预测模型。方法使用中国中医科学院望京医院骨科患者数据库作为数据来源;根据低蛋白血症诊断标准,选取数据库中的低蛋白血症患者,提取其人口学特征、生命体征、实验室检查等数据;使用逻辑回归、支持向量机、XGBoost(Extreme Gradient Boosting,极端梯度增强算法)等算法开发模型,并对不同机器学习算法在预测模型中使用效果的进行评估。结果共纳入526例患者,低蛋白血症发生率40.87%。逻辑回归模型的AUROC(Area Under the Receiver Operating Characteristic Curve,受试者工作特征曲线下面积)为0.8077,F值为0.853,整体性能优于随机森林、XGBoost等模型。同时模型性能比较结果显示,逻辑回归模型拥有较高的准确度。结论基于逻辑回归算法的高龄骨科患者术后低蛋白血症风险预测模型对辅助临床医生进行风险预测有重要意义。展开更多
Background:Clinical parameter-based nomograms and staging systems provide limited information for the prediction of survival in intrahepatic cholangiocarcinoma(ICC)patients.In this study,we developed a methylation sig...Background:Clinical parameter-based nomograms and staging systems provide limited information for the prediction of survival in intrahepatic cholangiocarcinoma(ICC)patients.In this study,we developed a methylation signature that precisely predicts overall survival(OS)after surgery.Methods:An epigenome-wide study of DNA methylation based on whole-genome bisulfite sequencing(WGBS)was conducted for two independent cohorts(discovery cohort,n=164;validation cohort,n=170)from three hepatobiliary centers in China.By referring to differentially methylated regions(DMRs),we proposed the concept of prognostically methylated regions(PMRs),which were composed of consecutive prognostically methylated CpGs(PMCs).Using machine learning strategies(Random Forest and the least absolute shrinkage and selector regression),a prognostic methylation score(PMS)was constructed based on 14 PMRs in the discovery cohort and confirmed in the validation cohort.Results:The C-indices of the PMS for predicting OS in the discovery and validation cohorts were 0.79 and 0.74,respectively.In the whole cohort,the PMS was an independent predictor of OS[hazard ratio(HR)=8.12;95% confidence interval(CI):5.48-12.04;P<0.001],and the C-index(0.78)of the PMS was significantly higher than that of the Johns Hopkins University School of Medicine(JHUSM)nomogram(0.69,P<0.001),the Eastern Hepatobiliary Surgery Hospital(EHBSH)nomogram(0.67,P<0.001),American Joint Committee on Cancer(AJCC)tumor-node-metastasis(TNM)staging system(0.61,P<0.001),and MEGNA prognostic score(0.60,P<0.001).The patients in quartile 4 of PMS could benefit from adjuvant therapy(AT)(HR=0.54;95%CI:0.32-0.91;log-rank P=0.043),whereas those in the quartiles 1-3 could not.However,other nomograms and staging system failed to do so.Further analyses of potential mechanisms showed that the PMS was associated with tumor biological behaviors,pathway activation,and immune microenvironment.Conclusions:The PMS could improve the prognostic accuracy and identify patients who would benefit from AT for ICC patients,and might facilitate decisions in treatment of ICC patients.展开更多
This retrospective analysis aimed to investigate the mutation profile of 16 common mutated genes in de novo acute myeloid leukemia (AML) patients. A total of 259 patients who were diagnosed of de novo AML were enrolle...This retrospective analysis aimed to investigate the mutation profile of 16 common mutated genes in de novo acute myeloid leukemia (AML) patients. A total of 259 patients who were diagnosed of de novo AML were enrolled in this study. Mutation profiling of 16 candidate genes were performed in bone marrow samples by using Sanger sequencing. We identified at least 1 mutation in 199 of the 259 samples (76.8%), and 2 or more mutations in 31.7% of samples. FLT3-ITD was the most common mutated gene (16.2%, 42/259), followed by CEBPA (15.1%, 39/259), NRAS (14.7%, 38/259), and NPM1 (13.5%, 35/259). Concurrence was observed in 97.1% of the NPM1 mutated cases and in 29.6% of the double mutated CEBPA cases. Distinct patterns of co-occurrence were observed for different hotspot mutations within the IDH2 gene: R140 mutations were associated with NPM1 and/or FLT3-ITD mutations, whereas R172 mutations co-occurred with DNMT3A mutations only. Concurrence was also observed in 86.6% of epigenetic regulation genes, most of which co-occurred with NPM1 mutations. The results showed certain rules in the mutation profiling and concurrence of AML patients, which was related to the function classification of genes. Defining the mutation spectrum and mutation pattern of AML will contribute to the comprehensive assessment of patients and identification of new therapeutic targets.展开更多
Single-tube bi-directional allele specific amp-lification(SB-ASA)and real-time quantitative polymer-ase chain reaction(RQ-PCR)assays were developed and performed for JAK2V617F detection on 40 polycythemia vera(PV)samp...Single-tube bi-directional allele specific amp-lification(SB-ASA)and real-time quantitative polymer-ase chain reaction(RQ-PCR)assays were developed and performed for JAK2V617F detection on 40 polycythemia vera(PV)samples,31 essential thrombocythemia(ET)samples,40 acute leukemia samples,and 40 healthy con-trol samples.Differences between detect limitations of the two assays and their influence on the mutation detection rate were analyzed.The results showed that in some sam-ples,the JAK2V617F burden was as low as nearly 1%,and thus more JAK2V617F-positive samples were detected by RQ-PCR than by SB-ASA assay due to the former higher detect limitation.Mutation allele ratios in PV and ET samples and their relevance to biological char-acteristics were also analyzed.The results showed that the mutation allele ratio was 0.436±0.261 in PV,higher than the 0.216±0.207 in ET;percentage of certainly homo-zygous mutation carriers in PV was 40.54%,higher than the 10%in ET.However,statistical analysis showed no relevance between mutation allele burden and sex or age.Our result shows that the pathogenesis of PV and ET may be related to the mutation allele burden of JAK2V617F.展开更多
文摘目的使用不同机器学习算法开发高龄骨科患者术后低蛋白血症风险预测模型。方法使用中国中医科学院望京医院骨科患者数据库作为数据来源;根据低蛋白血症诊断标准,选取数据库中的低蛋白血症患者,提取其人口学特征、生命体征、实验室检查等数据;使用逻辑回归、支持向量机、XGBoost(Extreme Gradient Boosting,极端梯度增强算法)等算法开发模型,并对不同机器学习算法在预测模型中使用效果的进行评估。结果共纳入526例患者,低蛋白血症发生率40.87%。逻辑回归模型的AUROC(Area Under the Receiver Operating Characteristic Curve,受试者工作特征曲线下面积)为0.8077,F值为0.853,整体性能优于随机森林、XGBoost等模型。同时模型性能比较结果显示,逻辑回归模型拥有较高的准确度。结论基于逻辑回归算法的高龄骨科患者术后低蛋白血症风险预测模型对辅助临床医生进行风险预测有重要意义。
基金supported by The National Key Technologies R&D Program of China(No.2018YFC1106800)The 1.3.5 project for disciplines of excellence,West China Hospital,Sichuan University(No.ZYJC18008)The Natural Science Foundation of China(Nos.91859105,81773012,81872004,81802302 and 81902401).
文摘Background:Clinical parameter-based nomograms and staging systems provide limited information for the prediction of survival in intrahepatic cholangiocarcinoma(ICC)patients.In this study,we developed a methylation signature that precisely predicts overall survival(OS)after surgery.Methods:An epigenome-wide study of DNA methylation based on whole-genome bisulfite sequencing(WGBS)was conducted for two independent cohorts(discovery cohort,n=164;validation cohort,n=170)from three hepatobiliary centers in China.By referring to differentially methylated regions(DMRs),we proposed the concept of prognostically methylated regions(PMRs),which were composed of consecutive prognostically methylated CpGs(PMCs).Using machine learning strategies(Random Forest and the least absolute shrinkage and selector regression),a prognostic methylation score(PMS)was constructed based on 14 PMRs in the discovery cohort and confirmed in the validation cohort.Results:The C-indices of the PMS for predicting OS in the discovery and validation cohorts were 0.79 and 0.74,respectively.In the whole cohort,the PMS was an independent predictor of OS[hazard ratio(HR)=8.12;95% confidence interval(CI):5.48-12.04;P<0.001],and the C-index(0.78)of the PMS was significantly higher than that of the Johns Hopkins University School of Medicine(JHUSM)nomogram(0.69,P<0.001),the Eastern Hepatobiliary Surgery Hospital(EHBSH)nomogram(0.67,P<0.001),American Joint Committee on Cancer(AJCC)tumor-node-metastasis(TNM)staging system(0.61,P<0.001),and MEGNA prognostic score(0.60,P<0.001).The patients in quartile 4 of PMS could benefit from adjuvant therapy(AT)(HR=0.54;95%CI:0.32-0.91;log-rank P=0.043),whereas those in the quartiles 1-3 could not.However,other nomograms and staging system failed to do so.Further analyses of potential mechanisms showed that the PMS was associated with tumor biological behaviors,pathway activation,and immune microenvironment.Conclusions:The PMS could improve the prognostic accuracy and identify patients who would benefit from AT for ICC patients,and might facilitate decisions in treatment of ICC patients.
文摘This retrospective analysis aimed to investigate the mutation profile of 16 common mutated genes in de novo acute myeloid leukemia (AML) patients. A total of 259 patients who were diagnosed of de novo AML were enrolled in this study. Mutation profiling of 16 candidate genes were performed in bone marrow samples by using Sanger sequencing. We identified at least 1 mutation in 199 of the 259 samples (76.8%), and 2 or more mutations in 31.7% of samples. FLT3-ITD was the most common mutated gene (16.2%, 42/259), followed by CEBPA (15.1%, 39/259), NRAS (14.7%, 38/259), and NPM1 (13.5%, 35/259). Concurrence was observed in 97.1% of the NPM1 mutated cases and in 29.6% of the double mutated CEBPA cases. Distinct patterns of co-occurrence were observed for different hotspot mutations within the IDH2 gene: R140 mutations were associated with NPM1 and/or FLT3-ITD mutations, whereas R172 mutations co-occurred with DNMT3A mutations only. Concurrence was also observed in 86.6% of epigenetic regulation genes, most of which co-occurred with NPM1 mutations. The results showed certain rules in the mutation profiling and concurrence of AML patients, which was related to the function classification of genes. Defining the mutation spectrum and mutation pattern of AML will contribute to the comprehensive assessment of patients and identification of new therapeutic targets.
基金supported by Key International Science and Technology Cooperation Projects of Ministry of Science and Technology of China(No.2006DFB31430)the National Natural Science Foundation of China(Grant No.30470939).
文摘Single-tube bi-directional allele specific amp-lification(SB-ASA)and real-time quantitative polymer-ase chain reaction(RQ-PCR)assays were developed and performed for JAK2V617F detection on 40 polycythemia vera(PV)samples,31 essential thrombocythemia(ET)samples,40 acute leukemia samples,and 40 healthy con-trol samples.Differences between detect limitations of the two assays and their influence on the mutation detection rate were analyzed.The results showed that in some sam-ples,the JAK2V617F burden was as low as nearly 1%,and thus more JAK2V617F-positive samples were detected by RQ-PCR than by SB-ASA assay due to the former higher detect limitation.Mutation allele ratios in PV and ET samples and their relevance to biological char-acteristics were also analyzed.The results showed that the mutation allele ratio was 0.436±0.261 in PV,higher than the 0.216±0.207 in ET;percentage of certainly homo-zygous mutation carriers in PV was 40.54%,higher than the 10%in ET.However,statistical analysis showed no relevance between mutation allele burden and sex or age.Our result shows that the pathogenesis of PV and ET may be related to the mutation allele burden of JAK2V617F.
