基于机器学习的肿瘤免疫治疗应答预测研究
|
摘要
肿瘤免疫治疗应答的预测对肿瘤治疗方案设计及治疗有着重要的意义。本文引入基于随机森林的机器学习方法,将病人黑色素瘤组织转录组RNA-seq的基因表达谱作为特征,对免疫检查点阻断治疗的结果进行预测研究。对病人的基因表达谱使用随机森林算法来构建预测模型,并与Logistic回归模型和XGBoost模型进行比较。实验结果表明,随机森林模型对免疫检查点阻断治疗的应答能够进行较准确的预测,并且较Logistic回归模型和XGBoost模型预测效果更好。
Prediction of the response to tumor immunotherapy is of great significance to the design of tumor treatment and treatment. In this paper, random forest machine learning method is introduced, and gene expression profile of patients' melanoma RNA-seq was taken as characteristics to predict the response to immune checkpoint blockade.Random forest algorithm was used to construct the prediction model for the gene expression profile of patients, and compared with Logistic regression analysis and XGBoost algorithm. The experimental results show that random forest model had a great prediction accuracy to the response to immune checkpoint blockade and was better than Logistic regression model and XGBoost model.
Prediction of the response to tumor immunotherapy is of great significance to the design of tumor treatment and treatment. In this paper, random forest machine learning method is introduced, and gene expression profile of patients' melanoma RNA-seq was taken as characteristics to predict the response to immune checkpoint blockade.Random forest algorithm was used to construct the prediction model for the gene expression profile of patients, and compared with Logistic regression analysis and XGBoost algorithm. The experimental results show that random forest model had a great prediction accuracy to the response to immune checkpoint blockade and was better than Logistic regression model and XGBoost model.
引文
[1]Stambrook PJ,Maher J,Farzaneh F.Cancer Immunotherapy:Whence and Whither[J].Mol Cancer Res.2017 Jun;15(6):635-650.
[2]卢伸,苏丹.免疫检查点阻断用于肿瘤治疗的研究进展[J].实用肿瘤杂志.2016;31(1):19-23.
[3]Topalian SL,et al.Mechanism-driven biomarkers to guide immune checkpoint blockade in cancer therapy[J].Nat.Rev.Cancer.2016 May;16(5):275-87.
[4]Liu XS,Mardis ER.Applications of immunogenomics to cancer[J].Cell.2017 Feb 9;168(4):600-612.
[5]Hugo W,Zaretsky JM,et al.Genomic and Transcriptomic Features of Response to Anti-PD-1 Therapy in Metastatic Melanoma[J].Cell.2016 Mar 24;165(1):35-44.
[6]Liu Q,et al.Towards In Silico Prediction of the ImmuneCheckpoint Blockade Response.[J].Trends Pharmacol Sci.2017 Dec;38(12):1041-1051.
[7]Galon J,et al.Type,density,and location of immune cells within human colorectal tumors predict clinical outcome.[J].Science.2006 Sep 29;313(5795):1960-4.
[8]Charoentong P,et al.Pan-cancer Immunogenomic Analyses Reveal Genotype-Immunophenotype Relationships and Predictors of Response to Checkpoint Blockade.Cell Rep.2017Jan 3;18(1):248-262.
[9]Cogdill AP,Andrews MC,Wargo JA.Hallmarks of response to immune checkpoint blockade[J].Br J Cancer.2017 Jun 27;117(1):1-7.
[10]李慧,李正,佘堃.一种基于综合不放回抽样的随机森林算法改进[J].计算机工程与科学.2015;7
[11]全雪峰.基于奇异熵和随机森林的人脸识别[J].软件,2016,37(02):35-38
[12]苏志同,汪武珺.基于随机森林的煅烧工艺参数的研究和分析[J].软件,2018,39(4):148-150
[13]Li Y,et al.A Mini-Review for Cancer Immunotherapy:Molecular Understanding of PD-1/PD-L1 Pathway Translational Blockade of Immune Checkpoints[J].Int J Mol Sci.2016 Jul18;17(7).pii:E1151.
[14]董师师,黄哲学.随机森林理论浅析[J].集成技术.2013.1;2(1):1-7.
[15]李欣海.随机森林模型在分类与回归分析中的应用[J].应用昆虫学报.2013,50(4):1190-1197.
[16]李玲,李晋宏.基于随机森林修正的加权二部图推荐算法[J].软件,2018,39(1):110-115.
[17]Riaz N,et al.Tumor and Microenvironment Evolution during Immunotherapy with Nivolumab[J].Cell.2017 Nov 2;171(4):934-949.e16.
[18]Tomczak K,Czerwińska P,Wiznerowicz M.The Cancer Genome Atlas(TCGA):an immeasurable source of knowledge[J].Contemp Oncol(Pozn).2015;19(1A):A68-77.
[19]吴荣强,李晋宏.基于聚类分析的铝电解槽阳极压降的分类[J].软件,2018,39(3):166-169.
[20]蒲杰方,卢荧玲.基于聚类算法和神经网络的客户分类模型构建[J].软件,2018,39(4):130-136.
[2]卢伸,苏丹.免疫检查点阻断用于肿瘤治疗的研究进展[J].实用肿瘤杂志.2016;31(1):19-23.
[3]Topalian SL,et al.Mechanism-driven biomarkers to guide immune checkpoint blockade in cancer therapy[J].Nat.Rev.Cancer.2016 May;16(5):275-87.
[4]Liu XS,Mardis ER.Applications of immunogenomics to cancer[J].Cell.2017 Feb 9;168(4):600-612.
[5]Hugo W,Zaretsky JM,et al.Genomic and Transcriptomic Features of Response to Anti-PD-1 Therapy in Metastatic Melanoma[J].Cell.2016 Mar 24;165(1):35-44.
[6]Liu Q,et al.Towards In Silico Prediction of the ImmuneCheckpoint Blockade Response.[J].Trends Pharmacol Sci.2017 Dec;38(12):1041-1051.
[7]Galon J,et al.Type,density,and location of immune cells within human colorectal tumors predict clinical outcome.[J].Science.2006 Sep 29;313(5795):1960-4.
[8]Charoentong P,et al.Pan-cancer Immunogenomic Analyses Reveal Genotype-Immunophenotype Relationships and Predictors of Response to Checkpoint Blockade.Cell Rep.2017Jan 3;18(1):248-262.
[9]Cogdill AP,Andrews MC,Wargo JA.Hallmarks of response to immune checkpoint blockade[J].Br J Cancer.2017 Jun 27;117(1):1-7.
[10]李慧,李正,佘堃.一种基于综合不放回抽样的随机森林算法改进[J].计算机工程与科学.2015;7
[11]全雪峰.基于奇异熵和随机森林的人脸识别[J].软件,2016,37(02):35-38
[12]苏志同,汪武珺.基于随机森林的煅烧工艺参数的研究和分析[J].软件,2018,39(4):148-150
[13]Li Y,et al.A Mini-Review for Cancer Immunotherapy:Molecular Understanding of PD-1/PD-L1 Pathway Translational Blockade of Immune Checkpoints[J].Int J Mol Sci.2016 Jul18;17(7).pii:E1151.
[14]董师师,黄哲学.随机森林理论浅析[J].集成技术.2013.1;2(1):1-7.
[15]李欣海.随机森林模型在分类与回归分析中的应用[J].应用昆虫学报.2013,50(4):1190-1197.
[16]李玲,李晋宏.基于随机森林修正的加权二部图推荐算法[J].软件,2018,39(1):110-115.
[17]Riaz N,et al.Tumor and Microenvironment Evolution during Immunotherapy with Nivolumab[J].Cell.2017 Nov 2;171(4):934-949.e16.
[18]Tomczak K,Czerwińska P,Wiznerowicz M.The Cancer Genome Atlas(TCGA):an immeasurable source of knowledge[J].Contemp Oncol(Pozn).2015;19(1A):A68-77.
[19]吴荣强,李晋宏.基于聚类分析的铝电解槽阳极压降的分类[J].软件,2018,39(3):166-169.
[20]蒲杰方,卢荧玲.基于聚类算法和神经网络的客户分类模型构建[J].软件,2018,39(4):130-136.