Abstract
目的
通过生物信息学方法，分析多种蛋白质联合预测结直肠癌(CRC)预后的作用及潜在的分子机制。
方法
从癌症蛋白质组图集(TCPA)数据库下载CRC蛋白质表达数据及临床数据，应用Perl及R软件对数据进行整理后筛选出预后相关的蛋白质; 进一步通过多因素Cox分析筛选出可作为CRC预后独立风险因子的蛋白质并据此构建预测模型。对模型中每一个蛋白质及模型风险评分进行生存分析，并对风险评分与患者生存状态绘制风险曲线验证预测模型对预后的预测作用。独立预后分析及ROC分析可反映预测模型在预后预测中的价值及优势。对模型蛋白质与CRC所有的相关蛋白质进行相互作用分析并在mRNA水平分析关键蛋白相关基因的差异表达。
结果
通过单因素及多因素Cox分析筛选出了6个蛋白质用于预测模型的构建; 生存分析发现与单个基因相比，预测模型表现出更强大的预后价值。单因素及多因素独立预后分析均提示预测模型风险评分与预后显著相关(P < 0.001)，预测模型可作为评估患者预后的独立风险因子; ROC分析显示预测模型在预后预测中表现出更加稳定的特异性和灵敏度(AUC=0.734)。蛋白质相互作用关系显示，蛋白质BID、SLC1A5及SRC_pY527与其他蛋白质表现出较明显的相关性(P < 0.001)，蛋白质SLC1A5及SRC_pY527与其他CRC相关的蛋白质间的相互作用最显著(SLC1A5与11种蛋白质间存在显著相关性; SRC_pY527与12种蛋白质间存在显著相关性，P < 0.001);除INPP4B外，各蛋白相关基因在mRNA水平均呈差异表达(P < 0.05)。
结论
6种蛋白质构建的预测模型对CRC 具有较好的预后预测作用，同时蛋白质SLC1A5及SRC_pY527在CRC的预后中起关键的作用，尤其是蛋白质SRC_pY527可能通过SRC/ AKT/MAPK信号轴调节CRC的发生发展并有望为CRC的治疗提供新的靶标。
Keywords: 结直肠癌, 预测模型, 预后, TCPA, 生物信息学分析
Abstract
Objective
To assess the value of the combination of multiple proteins in predicting the prognosis of colorectal cancer (CRC) through bioinformatics analysis.
Method
The protein expression and clinical data were downloaded from TCPA database. Perl and R were used to screen the prognostic-related proteins, and through Cox analysis, the proteins that served as independent prognostic factors of CRC were identified to build the prediction model. Survival analyses were conducted for each of the proteins included in the prediction model and the risk score of the model, and risk curves was drawn for the risk score and the patients' survival status to verify the performance of the model. Independent prognosis analysis and ROC analysis were used to assess the value and advantages of the model in prognosis prediction. The interactions between the proteins included in the model and the differential expressions of the key genes related with the proteins were analyzed.
Results
Six proteins were screened for model construction. Compared with a single gene, the model showed much greater prognostic value for CRC. Independent prognostic analysis showed that the risk score of the prediction model was significantly related with the prognosis (P < 0.001), and the model could be used as an independent risk factor for prognostic assessment of the patients. ROC analysis showed that the model had good specificity and sensitivity for prognostic prediction (AUC=0.734). Protein interactions showed that BID, SLC1A5 and SRC_pY527 were significantly correlated with other proteins (P < 0.001), and SLC1A5 and SRC_pY527 had the most significant interactions with other proteins (P < 0.001). Except for those of INPP4B, the key genes related with the proteins in the prediction model had significant differential expressions at the mRNA level in CRC (P < 0.05).
Conclusion
The prediction model constructed based on 6 proteins has good prognostic value for CRC. The proteins SLC1A5 and SRC_pY527 play key roles in the prognosis of CRC, and SRC_pY527 may regulate the occurrence and progression of CRC through the SRC/AKT/MAPK signal axis and thus may serve as a new therapeutic target of CRC.
Keywords: colorectal cancer, predictive model, prognosis, TCPA, bioinformatics analysis
结直肠癌(CRC)是胃肠道中常见的恶性肿瘤，我国以41~65岁人群发病率较高。近20年来，尤其在城市中，发病率明显上升[1]。70% CRC是由腺瘤性息肉病演变而来，从形态学可见到增生、腺瘤及癌变各阶段相应的染色体改变。随着分子生物学技术的发展，CRC癌变过程中的基因改变被逐渐认识。CRC的发生发展是一个多步骤、多阶段及多基因参与的细胞遗传性疾病[2]。近年来，关于CRC的基础研究主要集中在基因改变在CRC的诊断及治疗中的作用; 而关于预后的研究较少。
以往关于CRC预后的研究大多是基于其临床病理数据(例如肿瘤大小，肿瘤数量，淋巴结及血管浸润等)和单分子生物标记物(例如癌胚抗原CEA及糖类抗原CA199，CA125等)所作的预测[3-6]。然而，由于临床病理数据的收集存在较大的个体主观性及CRC发生发展过程的复杂性，单纯依靠临床病理数据或单个分子生物标记物均不能很好的对CRC的预后作出预测[7-8]。相关研究发现，在乳腺癌、前列腺癌及肝癌等肿瘤中联合多个基因所作的预后模型可显著提高预后预测的准确性[9-11]。蛋白表达基因在肿瘤发生的不同阶段均发挥作用，在结直肠癌中尚无相关研究。基于蛋白水平的研究相较于RNA水平更有利于在临床上的应用。基于此，本研究拟在蛋白层面对CRC进行分析，并据此从蛋白角度构建多蛋白分析模型; 以期提高对CRC预后预测的准确性，为CRC的临床诊断、治疗及预后提供重要的参考资料。
1. 材料和方法
1.1. 数据收集
从TCPA数据库下载CRC蛋白质表达数据(包含327个CRC样本及223种蛋白质)及临床数据(包含452个临床病例信息)<sup>[<xref ref-type="bibr" rid="b12">12</xref>]</sup>。从网站(<a href="https://www.perl.org/" target="_blank">https://www.perl.org/</a>)下载Perl软件并安装<sup>[<xref ref-type="bibr" rid="b13">13</xref>]</sup>;从网站(<a href="https://www.r-project.org/" target="_blank">https://www.r-project.org/</a>)下载R软件并安装<sup>[<xref ref-type="bibr" rid="b14">14</xref>]</sup>。
1.2. 构建蛋白质预测模型
通过Perl软件将蛋白质表达数据与患者生存数据进行整合，然后引用R软件“survival”包进行单因素Cox分析，并按P < 0.05的标准筛选筛选与CRC预后相关的蛋白质，引用R软件“ggplot2”及“ggrepel”包绘制火山图对结果进行可视化。然后，利用LASSO回归对预后相关蛋白质去除多重共线性，以防止模型基因间的过度拟合。最后，对预后相关蛋白质进行多因素Cox分析筛选CRC独立预后相关的蛋白质构建预测模型并以风险评分展示。风险评分=(蛋白质1的风险系数×蛋白质1的表达)+(蛋白质2的风险系数×蛋白质2的表达)+ …… +(蛋白质n的风险系数×蛋白质n的表达)[15]。
1.3. 生存分析
应用Perl软件将预测模型中的蛋白质表达数据与生存数据进行合并; 引用R软件“survival”包分别对蛋白质进行生存分析。然后，根据风险评分将患者分为高风险组(风险评分>中位风险评分)和低风险组(风险评分 < 中位风险评分)，并对风险评分进行生存分析。最后，对结果绘制生存曲线可视化。
1.4. 预测模型的预测价值验证
应用R软件“pheatmap”包根据风险评分对样品进行排序; 然后，对风险评分与蛋白质表达数据绘制风险热图分析模型蛋白质在高风险及低风险评分中的表达情况; 对风险评分与患者评分绘制风险曲线以评价预测模型在评估患者生存风险中的作用; 对风险评分与生存状态绘制生存状态图以评价模型对患者生存预后的预测作用[16]。
1.5. 独立预后分析
利用Perl软件将临床病理特征及风险评分与生存数据进行整合获得独立预后分析输入文件，利用R软件“survival”包对输入文件进行单因素及多因素cox回归分析以评价预测模型风险评分在预后预测中的作用。结果绘制森林图进行可视化。
1.6. ROC分析
以蛋白质预测模型风险评分与临床病理特征及患者生存状态作为输入数据，引用R软件“survivalROC”包对蛋白质预测模型风险评分及CRC临床病理数据进行ROC分析并绘制ROC曲线。AUC在0.5~0.7时有较低准确性，AUC在0.7~0.9时有一定准确性，AUC在0.9以上时有较高准确性[17]。
1.7. 蛋白质相关性分析
预测模型中蛋白质及CRC相关的223种蛋白质的表达数据作为输入数据，利用R软件进行相关性分析，并以相关系数cor>0.4及P < 0.001作为筛选条件对结果进行筛选。以相关性结果作为输入文件，引用R软件“ggplot2”和“ggalluvial”包绘制桑基图进行可视化。
1.8. 关键蛋白在mRNA水平的差异分析
从TCGA数据库CRC基因表达矩阵中提取关键蛋白的相关基因表达数据，并利用R软件的“Limma”及“beeswarm”程序包分析其在CRC组中的差异表达; 结果以散点图进行可视化。
1.9. 统计学方法
蛋白质表达数据的差异分析通过两独立样本的t检验完成; 生存分析组间比较的方法采用Kaplan-Meier方法; 采用LASSO回归分析去除多重共线性; 模型的构建、独立预后分析均通过多因素Cox分析完成; 相关性分析采用Pearson相关性检验。P < 0.05为差异具有统计学意义。
2. 结果
2.1. 构建蛋白质预测模型
单因素Cox分析显示，223个蛋白与CRC预后相关(P < 0.05，图 1A); LASSO回归去除共线性筛选出9个蛋白(图 1B、C); 多因素Cox分析共筛选获得6种蛋白质可作为CRC预后的独立风险因子(其中BID、SRC及SRC_pY527呈低风险; IGFBP2、INPP4B及SLC1A5呈高风险)，同时获得对应的风险系数(risk coefficient，coef)(表 1)。风险评分=(-1.7640 × BID的表达)+ (0.3286×IGFBP2的表达)+(0.5976×INPP4B的表达)+ (-1.0119×SRC的表达)+(-0.4386×SRC_pY527的表达)+(0.4698×SLC1A5的表达)。
1.
预后相关蛋白质及其LASSO回归分析
Prognosis-related proteins and their LASSO regression analysis. Red indicates that the high-risk proteins and green the low-risk ones.
1.
用于预测模型构建的蛋白质及其风险系数
Proteins and their risk coefficients used for construction of the prediction model
| ID | coef | HR | HR.95L | HR.95H | P |
| Note: Coef represents the risk factor. Risk score=(-1.7640 × BID expression) + (0.3286 × IGFBP2 expression) + (0.5976×INPP4B expression)+(-1.0119×SRC expression)+(-0.4386×SRC_pY527 expression)+(0.4698×SLC1A5 expression). | |||||
| BID | -1.7640 | 0.1714 | 0.0475 | 0.6187 | 0.0071 |
| IGFBP2 | 0.3286 | 1.3891 | 1.0486 | 1.8401 | 0.0220 |
| INPP4B | 0.5976 | 1.8177 | 1.1134 | 2.9674 | 0.0169 |
| SRC | -1.0119 | 0.3635 | 0.1894 | 0.6979 | 0.0024 |
| SRC_pY527 | -0.4386 | 0.6449 | 0.4527 | 0.9189 | 0.0152 |
| SLC1A5 | 0.4698 | 1.5997 | 0.8749 | 2.9249 | 0.1270 |
2.2. 生存分析
通过对6种模型蛋白质进行生存分析发现，蛋白质BID、SRC及SRC_pY527高表达患者预后良好(P < 0.05)，他们在CRC的发生发展中可能作为抑癌因子(图 2A、D、E); 而蛋白质IGFBP2、INPP4B及SLC1A5高表达患者预后不良(P < 0.05)，他们在CRC的发生发展中可能作为促癌因子(图 2B、C、F)。通过对预测模型风险评分进行生存分析(图 2G)，发现高风险组相较于低风险组总体生存较差(P < 0.001)。
2.
预测模型蛋白质及风险评分生存分析
Survival analysis of the proteins for building the prediction model and the risk score of the model. A: Survival rate of BID and CRC. B: Survival rate of IGFBP2 and CRC. C: Survival rate of INPP4B and CRC. D: Survival rate of SRC and CRC. E: Survival rate of SRC-pY527 and CRC. F: Survival rate of SLC1A5 and CRC. G: Survival rate of risk and CRC.
2.3. 预测模型对预后的预测价值验证
风险热图显示，蛋白质BID、SRC及SRC_pY527在高风险组呈低表达，IGFBP2、INPP4B及SLC1A5在高风险组呈高表达，这与生存分析结果具有相同的趋势(图 3C); 风险曲线显示，随着患者生存风险评分的递增其对应的模型风险评分相应增加(图 3A); 生存状态图显示，随着患者生存风险评分的增加患者生存率下降(图 3B)。
3.
预测模型风险曲线、生存状态图及风险热图
Predictive model risk curve (A), survival state diagram (B) and risk heat map (C).
2.4. 预测模型预测性能的验证
单因素(图 4A)和多因素(图 4B)Cox回归分析表明患者年龄和预测模型都是影响预后的独立危险因子(P < 0.05);ROC分析可检验预测模型对评估患者预后的准确性及灵敏度，通过对预测模型风险评分及临床病理特征进行ROC分析发现预测模型相较于传统的临床病理具有更高的准确性和灵敏度(图 4C)。
4.
预测模型独立预后及ROC分析
Independent prognosis and ROC analysis of the prediction model. A: Single factor independent prognostic analysis; B: Multivariate independent prognostic analysis; C: ROC analysis of clinicopathological and predictive models.
2.5. 蛋白质相关性分析
根据蛋白质相关性分析结果显示：蛋白质BID、SRC_pY527及SLC1A5与其他蛋白质具有较强的相关性，尤其是蛋白质SRC_pY527及SLC1A5(表 2，图 5A)。进一步对相关蛋白分析发现，它们大部分作为促癌蛋白并参与癌症通路AKT、MAPK及MEK等信号通路而促进肿瘤细胞的增殖、侵袭和转移; 蛋白质BID与癌细胞的凋亡相关，SRC_pY527与癌细胞的增殖相关，SLC1A5与癌细胞的细胞周期相关。
2.
蛋白质相关性分析结果
Protein correlation analysis results
| Protein1 | Protein2 | Cor | P |
| Note: Protein1 is the protein in the prediction model, protein2 is other CRC- related proteins; cor is the correlation coefficient between proteins. | |||
| BID | BAK | 0.526 | 1.21E-24 |
| BID | MRE11 | 0.595 | 9.50E-33 |
| BID | NCADHERIN | 0.534 | 1.69E-25 |
| SRC_pY527 | AKT_pS473 | 0.565 | 4.95E-29 |
| SRC_pY527 | GSK3ALPHABETA_pS21S9 | 0.733 | 3.08E-56 |
| SRC_pY527 | MAPK_pT202Y204 | 0.651 | 7.11E-41 |
| SRC_pY527 | MEK1_pS217S221 | 0.591 | 3.54E-32 |
| SRC_pY527 | P38_pT180Y182 | 0.635 | 2.35E-38 |
| SRC_pY527 | SHC_pY317 | 0.546 | 8.54E-27 |
| SRC_pY527 | SRC_pY416 | 0.534 | 1.87E-25 |
| SRC_pY527 | YB1_pS102 | 0.523 | 2.53E-24 |
| SRC_pY527 | GSK3_pS9 | 0.684 | 1.89E-46 |
| SRC_pY527 | NDRG1_pT346 | 0.545 | 1.15E-26 |
| SRC_pY527 | TUBERIN_pT1462 | 0.534 | 1.82E-25 |
| SRC_pY527 | SHP2_pY542 | 0.609 | 1.38E-34 |
| SLC1A5 | BETACATENIN | 0.575 | 3.20E-30 |
| SLC1A5 | CYCLINB1 | 0.517 | 1.05E-23 |
| SLC1A5 | BRAF | 0.529 | 5.21E-25 |
| SLC1A5 | BAP1C4 | 0.589 | 7.24E-32 |
| SLC1A5 | EIF4G | 0.522 | 3.08E-24 |
| SLC1A5 | FOXM1 | 0.542 | 2.61E-26 |
| SLC1A5 | RBM15 | 0.504 | 2.00E-22 |
| SLC1A5 | GCN5L2 | 0.524 | 1.82E-24 |
| SLC1A5 | MSH6 | 0.542 | 2.18E-26 |
| SLC1A5 | BRD4 | 0.609 | 1.42E-34 |
| SLC1A5 | COG3 | 0.549 | 3.58E-27 |
5.
蛋白质相关性分析及模型蛋白在mRNA水平的差异表达
Protein correlation and the difference analysis of model-related proteins at the mRNA level. A: Correlation analysis showed that there was a correlation between BID and 3 proteins, SRC_pY527 was correlated with 12 proteins, and SLC1A5 was correlated with 11 proteins. B: Gene BID has a significant high expression in colorectal cancer; C: Gene IGFBP2 has a certain level of high expression in colorectal cancer; D: Gene INPP4B has no significant difference in expression in colorectal cancer; E: Gene SRC is in colorectal cancer Rectal cancer is significantly high expression; F: Gene SRC_pY527 is significantly high expression in colorectal cancer; G: gene SLC1A5 is significantly high expression in colorectal cancer.
3. 讨论
随着我国人口老龄化的加剧，CRC发病率在我国呈现上升趋势。CRC的治疗仍然是以手术切除及化学药物治疗为主的综合治疗。目前，由于对CRC缺乏早期的诊断手段，大多数患者被诊断时已属晚期而失去手术等治愈性治疗的时机且癌症进展过程中癌基因的突变使得化学治疗出现耐药导致CRC的预后较差[18]。虽然对CRC的诊断及治疗的研究较多，但均未能取得较大的突破。因此，寻找一种可靠的方法用于CRC预后的预测以及时准确的评估治疗效果并指导进一步治疗至关重要。
本研究通过对TCPA数据库中CRC相关的蛋白质进行分析构建了一种包含6种蛋白质的预测模型发现：该预测模型可以有效地对生存进行分层，并且高风险组相较于低风险组总体生存较差。独立预后及ROC分析提示：预测模型可作为患者预后的独立风险因子，并且对患者预后具有较好的预测价值。此外，通过对蛋白质相关性分析发现：BID、SRC_pY527及SLC1A5是调节CRC发生发展的关键蛋白质，它们与细胞凋亡(BAK、MEK1_pS217S221及FOXM1)、细胞增殖(AKT_pS473、SHC_pY317及SRC_pY416等)、细胞侵袭、转移(NCADHERIN、NDRG1_pT346及TUBERIN_pT1462)及肿瘤代谢(GSK3ALPHABETA_pS21S9)相关并通过AKT (AKT_pS473) 及MAPK (MAPK_pT202Y204、P38_pT180Y182及P38_pT180Y182)等信号通路发挥作用。总之，这些结果证明了该预后模型的巨大预后价值，同时发现这些蛋白质可能通过SRC/ AKT/MAPK信号轴调节CRC发生发展。
BID蛋白是Bcl-2家族中促凋亡类的蛋白。它具有可被caspase8酶切调控、高效的诱导细胞色素c从线粒体泄漏到细胞浆中的功能，从而在细胞凋亡中起重要作用[19]。BID蛋白还可以与Bax蛋白协同作用，通过促进Bax与线粒体的结合及引起Bax构象的变化，而加强Bax引起的线粒体损伤[20]。胰岛素样生长因子结合蛋白2(IGFBP2)，可通过增强基质金属蛋白2对细胞外基质的讲解并介导胰岛素样生长因子介导的信号转导过程，从而促进细胞增殖[21]。II型多磷酸肌醇4-磷酸酶(INPP4B)是一种抑癌基因，通过抑制AKT激酶，阻断PI3K/AKT信号转导通路，减弱肿瘤细胞的生长、增殖能力，诱发肿瘤凋亡[22-23]。SRC蛋白可激活Ras蛋白，然后依次激活Raf、MEK和MAPK/ERK，从而引发一系列的生物学效应; 同时活化的MAPK途径也可以磷酸化Src相应的位点，导致基因转录的激活，而抑制MAPK途径的活化可逆转Src蛋白的部分效应[24-25]。SRC蛋白异常激活可促进细胞增殖、侵袭转移及诱导血管生成等，并与胃癌、结直肠癌及肝癌等多种肿瘤的发生密切相关[26-27]。溶质载体家族1成员5(SLC1A5)，是氨基酸转运载体家族中的重要一员。主要转运包括谷氨酰胺在内的多种小分子中性氨基酸[28]。相关研究表明：SLC1A5在多种恶性肿瘤组织和细胞中高表达，并且与肿瘤增殖、侵袭及预后不良有关，体内外研究表明抑制SLC1A5可抑制肿瘤细胞生长[29-30]。这些蛋白质均与肿瘤的发生发展有关，并且在CRC中相关研究不足。
我们的研究集中于CRC中不断改变的蛋白质的预后作用，而不仅仅局限于单个蛋白质。该分析结果具有潜在的实质性临床意义，有望成为评估CRC治疗预后新的指标; 并对CRC的实验研究提供一个重要的研究方向，有望为CRC的诊治、治疗提供新的靶标。但是，尽管我们对本研究做了细致严格的分析，但是仍存在几个问题。第一，在构建蛋白质预测模型时，只有6个蛋白质被用于预测模型的构建，导致一些重要的蛋白质在构建模型之前已经被排除在外，并最终降低了预测模型的性能。第二，CRC的发生发展是一个多因素、多机制共同作用的复杂过程; 仅仅利用蛋白质构建预测模型来评估CRC的预后将导致预测性能的不足。第三，功能实验是必要的，以揭示模型相关蛋白质在调节CRC进展中的功能及SRC/ AKT/MAPK信号轴调节CRC发生发展分子机制。
Biography
温贺新，在读硕士研究生，E-mail: wenhexin66@126.com
Funding Statement
国家自然科学基金(82070561，81902078)；蚌埠医学院科研创新团队(BYKC201909)；蚌埠医学院研究生科研创新计划(Byycx20091)
Supported by National Natural Science Foundation of China (82070561, 81902078)
Contributor Information
温 贺新 (Hexin WEN), Email: wenhexin66@126.com.
左 芦根 (Lugen ZUO), Email: zuolugen@126.com.
刘 牧林 (Mulin LIU), Email: liumulin66@aliyun.com.
References
- 1.Siegel RL, Miller KD, Goding Sauer A, et al. Colorectal cancer statistics, 2020. CACancer J Clin. 2020;70(3):145–64. doi: 10.3322/caac.21601. [Siegel RL, Miller KD, Goding Sauer A, et al. Colorectal cancer statistics, 2020[J]. CACancer J Clin, 2020, 70(3): 145-64.] [DOI] [PubMed] [Google Scholar]
- 2.Curtius K, Wright NA, Graham TA. An evolutionary perspective on field cancerization. Nat Rev Cancer. 2018;18(1):19–32. doi: 10.1038/nrc.2017.102. [Curtius K, Wright NA, Graham TA. An evolutionary perspective on field cancerization[J]. Nat Rev Cancer, 2018, 18(1): 19-32.] [DOI] [PubMed] [Google Scholar]
- 3.Song YF, Huang ZY, Kang YL, et al. Clinical usefulness and prognostic value of red cell distribution width in colorectal cancer. Biomed Res Int. 2018;2018:9858943. doi: 10.1155/2018/9858943. [Song YF, Huang ZY, Kang YL, et al. Clinical usefulness and prognostic value of red cell distribution width in colorectal cancer [J]. Biomed Res Int, 2018, 2018: 9858943.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Ren F, Weng W, Zhang Q, et al. Clinicopathological features and prognosis of AFP-producing colorectal cancer: a single-center analysis of 20 cases. Cancer Manag Res. 2019;11:4557–67. doi: 10.2147/CMAR.S196919. [Ren F, Weng W, Zhang Q, et al. Clinicopathological features and prognosis of AFP-producing colorectal cancer: a single-center analysis of 20 cases[J]. Cancer Manag Res, 2019, 11: 4557-67.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Chen L, Jiang B, Di J, et al. Predictive value of preoperative detection of CEA and CA199 for prognosis in patients with stage ⅢⅡ colorectal cancer. http://europepmc.org/abstract/MED/26404690. Zhonghua Wei Chang Wai Ke Za Zhi. 2015;18(9):914–9. [Chen L, Jiang B, Di J, et al. Predictive value of preoperative detection of CEA and CA199 for prognosis in patients with stage ⅢⅡ colorectal cancer[J]. Zhonghua Wei Chang Wai Ke Za Zhi, 2015, 18(9): 914-9.] [PubMed] [Google Scholar]
- 6.Sun X, Huang T, Cheng F, et al. Monitoring colorectal cancer following surgery using plasma circulating tumor DNA. Oncol Lett. 2018;15(4):4365–75. doi: 10.3892/ol.2018.7837. [Sun X, Huang T, Cheng F, et al. Monitoring colorectal cancer following surgery using plasma circulating tumor DNA[J]. Oncol Lett, 2018, 15(4): 4365-75.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Wu GZ, Zhang ML. A novel risk score model based on eight genes and a nomogram for predicting overall survival of patients with osteosarcoma. BMC Cancer. 2020;20(1):456. doi: 10.1186/s12885-020-06741-4. [Wu GZ, Zhang ML. A novel risk score model based on eight genes and a nomogram for predicting overall survival of patients with osteosarcoma[J]. BMC Cancer, 2020, 20(1): 456.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Al-Ansari MM, Al-Saif M, Arafah M, et al. Clinical and functional significance of tumor/stromal ATR expression in breast cancer patients. Breast Cancer Res. 2020;22(1):49. doi: 10.1186/s13058-020-01289-4. [Al-Ansari MM, Al-Saif M, Arafah M, et al. Clinical and functional significance of tumor/stromal ATR expression in breast cancer patients[J]. Breast Cancer Res, 2020, 22(1): 49.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Tang CZ, Ma JK, Liu XL, et al. Identification of a prognostic signature of nine metabolism-related genes for hepatocellular carcinoma. PeerJ. 2020;8:e9774. doi: 10.7717/peerj.9774. [Tang CZ, Ma JK, Liu XL, et al. Identification of a prognostic signature of nine metabolism-related genes for hepatocellular carcinoma[J]. PeerJ, 2020, 8: e9774.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Liu Z, Mi M, Li X, et al. A lncRNA prognostic signature associated with immune infiltration and tumour mutation burden in breast cancer. J Cell Mol Med. 2020;24(21):12444–56. doi: 10.1111/jcmm.15762. [Liu Z, Mi M, Li X, et al. A lncRNA prognostic signature associated with immune infiltration and tumour mutation burden in breast cancer[J]. J Cell Mol Med, 2020, 24(21): 12444-56.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Zhou SY, Yan YL, Chen X, et al. A two-gene-based prognostic signature for pancreatic cancer. Aging. 2020;12(18):18322–42. doi: 10.18632/aging.103698. [Zhou SY, Yan YL, Chen X, et al. A two-gene-based prognostic signature for pancreatic cancer[J]. Aging, 2020, 12(18): 18322-42.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Chen MM, Li J, Wang Y, et al. TCPA v3.0: an integrative platform to explore the Pan-cancer analysis of functional proteomic data. http://www.ncbi.nlm.nih.gov/pubmed/31201206. Mol Cell Proteomics. 2019;18(8 suppl 1):S15–25. doi: 10.1074/mcp.RA118.001260. [Chen MM, Li J, Wang Y, et al. TCPA v3.0: an integrative platform to explore the Pan-cancer analysis of functional proteomic data[J]. Mol Cell Proteomics, 2019, 18(8 suppl 1): S15-25.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Bonnal RJP, Yates A, Goto N, et al. Sharing programming resources between bio* projects. Methods Mol Biol Clifton N J. 2019;1910:747–66. doi: 10.1007/978-1-4939-9074-0_25. [Bonnal RJP, Yates A, Goto N, et al. Sharing programming resources between bio* projects[J]. Methods Mol Biol Clifton N J, 2019, 1910: 747-66.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Shim SR, Kim SJ. Intervention meta-analysis: application and practice using R software. Epidemiol Health. 2019;41:e2019008. doi: 10.4178/epih.e2019008. [Shim SR, Kim SJ. Intervention meta-analysis: application and practice using R software[J]. Epidemiol Health, 2019, 41: e2019008.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Ding CG, Li Y, Tian XH, et al. Predictive score model for delayed graft function based on hypothermic machine perfusion variables in kidney transplantation. Chin Med J (Engl) 2018;131(22):2651–7. doi: 10.4103/0366-6999.245278. [Ding CG, Li Y, Tian XH, et al. Predictive score model for delayed graft function based on hypothermic machine perfusion variables in kidney transplantation[J]. Chin Med J (Engl), 2018, 131(22): 2651-7.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Jin J, Zhou S, Xu Q, et al. Identification of risk factors in epidemiologic study based on ROC curve and network. Sci Rep. 2017;7:46655. doi: 10.1038/srep46655. [Jin J, Zhou S, Xu Q, et al. Identification of risk factors in epidemiologic study based on ROC curve and network[J]. Sci Rep, 2017, 7: 46655.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Kamarudin AN, Cox T, Kolamunnage-Dona R. Time-dependent ROC curve analysis in medical research: current methods and applications. BMC Med Res Methodol. 2017;17(1):53. doi: 10.1186/s12874-017-0332-6. [Kamarudin AN, Cox T, Kolamunnage-Dona R. Time-dependent ROC curve analysis in medical research: current methods and applications[J]. BMC Med Res Methodol, 2017, 17(1): 53.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Zhang Y, Chen Z, Li J. The current status of treatment for colorectal cancer in China: a systematic review. Medicine: Baltimore. 2017;96(40):e8242. doi: 10.1097/MD.0000000000008242. [Zhang Y, Chen Z, Li J. The current status of treatment for colorectal cancer in China: a systematic review[J]. Medicine: Baltimore, 2017, 96(40): e8242.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Carlström KE, Zhu K, Ewing E, et al. Gsta4 controls apoptosis of differentiating adult oligodendrocytes during homeostasis and remyelination via the mitochondria-associated Fas-Casp8-Bid-axis. Nat Commun. 2020;11(1):4071. doi: 10.1038/s41467-020-17871-5. [Carlström KE, Zhu K, Ewing E, et al. Gsta4 controls apoptosis of differentiating adult oligodendrocytes during homeostasis and remyelination via the mitochondria-associated Fas-Casp8-Bid-axis [J]. Nat Commun, 2020, 11(1): 4071.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Debernardi J, Hollville E, Lipinski M, et al. Differential role of FLBID and -BID during verotoxin-1-induced apoptosis in Burkitt's lymphoma cells. Oncogene. 2018;37(18):2410–21. doi: 10.1038/s41388-018-0123-5. [Debernardi J, Hollville E, Lipinski M, et al. Differential role of FLBID and -BID during verotoxin-1-induced apoptosis in Burkitt's lymphoma cells[J]. Oncogene, 2018, 37(18): 2410-21.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Shen F, Song C, Liu Y, et al. IGFBP2 promotes neural stem cell maintenance and proliferation differentially associated with glioblastoma subtypes. Brain Res. 2019;1704:174–86. doi: 10.1016/j.brainres.2018.10.018. [Shen F, Song C, Liu Y, et al. IGFBP2 promotes neural stem cell maintenance and proliferation differentially associated with glioblastoma subtypes[J]. Brain Res, 2019, 1704: 174-86.] [DOI] [PubMed] [Google Scholar]
- 22.Chen Y, Sun Z, Qi M, et al. INPP4B restrains cell proliferation and metastasis via regulation of the PI3K/AKT/SGK pathway. J Cell Mol Med. 2018;22(5):2935–43. doi: 10.1111/jcmm.13595. [Chen Y, Sun Z, Qi M, et al. INPP4B restrains cell proliferation and metastasis via regulation of the PI3K/AKT/SGK pathway[J]. J Cell Mol Med, 2018, 22(5): 2935-43.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Zhao S, Wu LS, Kuang YH, et al. Downregulation of CD147 induces malignant melanoma cell apoptosis via the regulation of IGFBP2 expression. http://www.ncbi.nlm.nih.gov/pubmed/30272281. Int J Oncol. 2018;53(6):2397–408. doi: 10.3892/ijo.2018.4579. [Zhao S, Wu LS, Kuang YH, et al. Downregulation of CD147 induces malignant melanoma cell apoptosis via the regulation of IGFBP2 expression[J]. Int J Oncol, 2018, 53(6): 2397-408.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Man J, Cui K, Fu X, et al. Donepezil promotes neurogenesis via Src signaling pathway in a rat model of chronic cerebral hypoperfusion. Brain Res. 2020;1736:146782. doi: 10.1016/j.brainres.2020.146782. [Man J, Cui K, Fu X, et al. Donepezil promotes neurogenesis via Src signaling pathway in a rat model of chronic cerebral hypoperfusion [J]. Brain Res, 2020, 1736: 146782.] [DOI] [PubMed] [Google Scholar]
- 25.Wen W, Han ES, Dellinger TH, Lu LX, Wu J, Jove R, Yim JH. Synergistic anti-tumor activity by targeting multiple signaling pathways in ovarian cancer. Cancers (Basel) 2020;12(9):2586. doi: 10.3390/cancers12092586. [Wen W, Han ES, Dellinger TH, Lu LX, Wu J, Jove R, Yim JH. Synergistic anti-tumor activity by targeting multiple signaling pathways in ovarian cancer[J]. Cancers (Basel), 2020, 12(9): 2586.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Mengardo V, Treppiedi E, Bencivenga M, et al. Tailored treatment for signet ring cell gastric cancer. Updates Surg. 2018;70(2):167–71. doi: 10.1007/s13304-018-0550-4. [Mengardo V, Treppiedi E, Bencivenga M, et al. Tailored treatment for signet ring cell gastric cancer[J]. Updates Surg, 2018, 70(2): 167- 71.] [DOI] [PubMed] [Google Scholar]
- 27.Asiri A, Toss MS, Raposo TP, et al. Cten promotes EpithelialMesenchymal Transition (EMT) in colorectal cancer through stabilisation of Src. Pathol Int. 2019;69(7):381–91. doi: 10.1111/pin.12811. [Asiri A, Toss MS, Raposo TP, et al. Cten promotes EpithelialMesenchymal Transition (EMT) in colorectal cancer through stabilisation of Src[J]. Pathol Int, 2019, 69(7): 381-91.] [DOI] [PubMed] [Google Scholar]
- 28.Ghaffari MH, Sadri H, Hammon HM, et al. Short communication: Colostrum versus formula: Effects on mRNA expression of genes related to branched- chain amino acid metabolism in neonatal dairy calves. J Dairy Sci. 2020;103(10):9656–66. doi: 10.3168/jds.2020-18429. [Ghaffari MH, Sadri H, Hammon HM, et al. Short communication: Colostrum versus formula: Effects on mRNA expression of genes related to branched- chain amino acid metabolism in neonatal dairy calves[J]. J Dairy Sci, 2020, 103(10): 9656-66.] [DOI] [PubMed] [Google Scholar]
- 29.Osman I, He XQ, Liu JH, et al. TEAD1 (TEA domain transcription factor 1) promotes smooth muscle cell proliferation through upregulating SLC1A5 (solute carrier family 1 member 5)-mediated glutamine uptake. Circ Res. 2019;124(9):1309–22. doi: 10.1161/CIRCRESAHA.118.314187. [Osman I, He XQ, Liu JH, et al. TEAD1 (TEA domain transcription factor 1) promotes smooth muscle cell proliferation through upregulating SLC1A5 (solute carrier family 1 member 5)-mediated glutamine uptake[J]. Circ Res, 2019, 124(9): 1309-22.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Liu Y, Zhao T, Li Z, et al. The role of ASCT2 in cancer: a review. Eur J Pharmacol. 2018;837:81–7. doi: 10.1016/j.ejphar.2018.07.007. [Liu Y, Zhao T, Li Z, et al. The role of ASCT2 in cancer: a review[J]. Eur J Pharmacol, 2018, 837: 81-7.] [DOI] [PubMed] [Google Scholar]