0．1Instalation

if (!require("BiocManager")) install.packages("BiocManager")

1需要包

库(dplyr)库(ggplot2)库(survival)库(futile.logger)库(curatedTCGAData)库(TCGAutils)库(glmSparseNet) # #无用的一些通用选项。记录调试包。最后。值<- flag .layout(布局。format('[~l] ~m'))。value <- glmSparseNet:::show.message(FALSE) #设置ggplot2默认主题为最小theme_set(ggplot2::theme_minimal())

2加载数据

数据是从TCGA下载的在线精选数据集中加载的curatedTCGAData生物导体包装和加工。

为了加速这个过程，我们使用了一个非常精简的数据集，只有107个变量(基因)，它作为数据对象存储在这个包中。但是，下面的数据块描述了手动获取数据的过程。

brca <- curatedTCGAData(disease ecode = " brca "， assays = "RNASeq2GeneNorm"， version = "1.1.38"， dry.run = FALSE)

#只保留实体肿瘤(代码:01)brca.primary.solid.tumor <- TCGAutils::TCGAsplitAssays(brca， '01') xdata。raw <- t(assay(brca.primary.solid.tumor[[1]])) #获取生存信息。raw <- colData(brca.primary.solid.tumor) %>% as.data.frame %>% #只保留与存活或样本相关的数据dplyr::select(patientID, vitital_status, Days.to.date.of. of.)死亡,Days.to.Date.of.Last。联系人，days_to_death, days_to_last_followup, Vital.Status) %>% #将天数转换为整数dplyr::mutate(days .to.date.of. status) %>% #死亡= as.integer(days .to.date.of.Death)) %>% dplyr::mutate(days .to. last . contact = as.integer(days .to.date.of. last . contact)) %>% #查找所有天之间的最大时间(忽略缺失)dplyr::rowwise() %>% dplyr::mutate(time = max(days_to_last_followup, days .to.date.of. date.of. date. date. date.of. date. date. date. contact) %>% dplyr::mutate(time = max(days_to_last_followup, days .to.date.of. date. contact) %>%死亡,Days.to.Last。联系人，days_to_death, narm = TRUE)) %>% #只保留生存变量和代码dplyr::select(patientID, status = vitital_status, time) %>% #丢弃生存时间小于或等于0的个体dplyr::filter(!is.na(time) & time > 0) %>% as.data.frame() #将索引设置为patientID的行名(ydata.raw) <- ydata. frame)。获取生存数据和分析数据xdata之间的匹配。Raw <- xdata。raw[TCGAbarcode(rownames(xdata.raw)) %in% rownames(ydata.raw),] xdata.raw <- xdata.raw %>% { (apply(., 2, sd) != 0) } %>% { xdata.raw[, .] } %>% scale # Order ydata the same as assay ydata.raw <- ydata.raw[TCGAbarcode(rownames(xdata.raw)), ] # Using only a subset of genes previously selected to keep this short example. set.seed(params$seed) small.subset <- c('CD5', 'CSF2RB', 'IRGC', 'NEUROG2', 'NLRC4', 'PDE11A', 'PTEN', 'TP53', 'BRAF', 'PIK3CB', 'QARS', 'RFC3', 'RPGRIP1L', 'SDC1', 'TMEM31', 'YME1L1', 'ZBTB11', sample(colnames(xdata.raw), 100)) %>% unique xdata <- xdata.raw[, small.subset[small.subset %in% colnames(xdata.raw)]] ydata <- ydata.raw %>% dplyr::select(time, status)

3.合适的模型

拟合模型模型的惩罚由集线器利用交叉验证功能进行cv.glmHub．

Set.seed (params$seed)符合<- cv。glmHub(xdata, Surv(ydata$time, ydata$status)， family = 'cox'， lambda = buildLambda(1)，网络= 'correlation'，网络。选项= networkOptions(cutoff = .6, min.degree = .2))

##警告在正则化。值(x, y, ties, missing(ties)， na。rm = na.rm): ##崩溃为唯一的'x'值##正则化中的警告。值(x, y, ties, missing(ties)， na。rm = na.rm): ##崩溃为唯一的'x'值##正则化中的警告。值(x, y, ties, missing(ties)， na。rm = na.rm): ##崩溃为唯一的'x'值##正则化中的警告。值(x, y, ties, missing(ties)， na。rm = na.rm): ##崩溃为唯一的'x'值##正则化中的警告。值(x, y, ties, missing(ties)， na。rm = na.rm): ##崩溃为唯一的'x'值##正则化中的警告。值(x, y, ties, missing(ties)， na。rm = na.rm): ##崩溃为唯一的'x'值##正则化中的警告。值(x, y, ties, missing(ties)， na。rm = na.rm): ##崩溃为唯一的'x'值##正则化中的警告。值(x, y, ties, missing(ties)， na。rm = na.rm): ##崩溃为唯一的'x'值##正则化中的警告。值(x, y, ties, missing(ties)， na。rm = na.rm): ## collapsing to unique 'x' values ## Warning in regularize.values(x, y, ties, missing(ties), na.rm = na.rm): ## collapsing to unique 'x' values

4交叉验证结果

显示的结果One hundred.使用不同的参数在10倍交叉验证中找到最优值。两条垂直虚线表示最佳模型和选择的变量较少的模型(基因)，但与最佳结果在标准误差范围内。

情节(安装)

系数。v < -系数(安装、s = lambda.min) [1 ] %>% { .[.! = 0]}

##警告在正则化。值(x, y, ties, missing(ties)， na。Rm = na.rm): ##崩溃为唯一的'x'值

系数。v %>% {data.frame(gene.name = names(.)， coefficient = .， stringsAsFactors = FALSE)} %>% arrange(gene.name) %>% knitr::kable()

名称(coefs.v) %>%{标记(.)$热图}

## curl::curl_fetch_memory(url, handle = handle): Timeout was reached: [chat.lionproject.net]操作在10001毫秒后超时，接收到0个字节中的0个字节再试1.6秒…

## curl::curl_fetch_memory(url, handle = handle): OpenSSL SSL_connect: SSL_ERROR_SYSCALL in connection to chat.lionproject.net:443 ##请求失败[Error]。3秒后重试…

无法调用Hallmark API，请稍后再试。

# #空

separate2GroupsCox(as.vector(coefs.v)， xdata[， names(coefs.v)]， ydata, plot。title = '完整数据集'，图例。outside = FALSE)

## $pvalue ## [1] 0.001237802 ## ## $plot

## ## $km ##调用:survfit(公式=生存::Surv(时间，状态)~组，数据=预后。index.df) ## ## n事件中位数0.95LCL 0.95UCL ##低风险540 58 3959 3492 NA ##高风险540 94 3738 3262 4456

5会话信息

sessionInfo ()

## R版本4.2.0 RC (2022-04-21 r82226) ##平台:x86_64-pc-linux-gnu(64位)##运行在Ubuntu 20.04.4 LTS ## ##矩阵产品:默认## BLAS: /home/biocbuild/bbs-3.16-bioc/R/lib/libRblas。/home/biocbuild/bbs-3.16-bioc/R/lib/libRlapack。所以## ## locale: ## [1] LC_CTYPE=en_US。UTF-8 LC_NUMERIC= c# # [3] LC_TIME=en_GB LC_COLLATE= c# # [5] LC_MONETARY=en_US。utf - 8 LC_MESSAGES = en_US。UTF-8 ## [7] LC_PAPER=en_US。UTF-8 LC_NAME= c# # [9] LC_ADDRESS=C lc_phone = c# # [11] LC_MEASUREMENT=en_US。UTF-8 LC_IDENTIFICATION=C ## ##附加的基本包:##[1]网格并行stats4统计图形grDevices utils ##[8]数据集方法基础## ##其他附加包:# # # # [1] VennDiagram_1.7.3 reshape2_1.4.4 [3] forcats_0.5.1 glmSparseNet_1.15.0 # # [5] glmnet_4.1-4 Matrix_1.4-1 # # [7] TCGAutils_1.17.0 curatedTCGAData_1.17.1 # # [9] MultiAssayExperiment_1.23.0 SummarizedExperiment_1.27.0 # # [11] Biobase_2.57.0 GenomicRanges_1.49.0 # # [13] GenomeInfoDb_1.33.0 IRanges_2.31.0 # # [15] S4Vectors_0.35.0 BiocGenerics_0.43.0 # # [17] MatrixGenerics_1.9.0 matrixStats_0.62.0 # # [19] futile.logger_1.4.3 survival_3.3-1 # # [21] ggplot2_3.3.5 dplyr_1.0.8 # # [23]BiocStyle_2.25.0 ## ##通过命名空间加载(并且没有附加):# # # # [1] backports_1.4.1 AnnotationHub_3.5.0 [3] BiocFileCache_2.5.0 plyr_1.8.7 # # [5] splines_4.2.0 BiocParallel_1.31.0 # # [7] digest_0.6.29 foreach_1.5.2 # # [9] htmltools_0.5.2 magick_2.7.3 # # [11] fansi_1.0.3 magrittr_2.0.3 # # [13] memoise_2.0.1 tzdb_0.3.0 # # [15] Biostrings_2.65.0 readr_2.1.2 # # [17] prettyunits_1.1.1 colorspace_2.0-3 # # [19] blob_1.2.3 rvest_1.0.2 # # [21] rappdirs_0.3.3 xfun_0.30 # # [23] crayon_1.5.1 rcurl_1.98 - 1.6 # # [25] jsonlite_1.8.0 zoo_1.8-10 # # [27] iterators_1.0.14glue_1.6.2 ## [29] survminer_0.4.9 GenomicDataCommons_1.21.0 ## [31] gtable_0.3.0 zlibbioc_1.43.0 ## [33] XVector_0.37.0 DelayedArray_0.23.0 ## [35] car_3.0-12 shape_1.4.6 ## [37] abind_1.4-5 scales_1.2.0 ## [39] futile.options_1.0.1 DBI_1.1.2 ## [41] rstatix_0.7.0 Rcpp_1.0.8.3 ## [43] xtable_1.8-4 progress_1.2.2 ## [45] bit_4.0.4 km.ci_0.5-6 ## [47] httr_1.4.2 ellipsis_0.3.2 ## [49] pkgconfig_2.0.3 XML_3.99-0.9 ## [51] farver_2.1.0 sass_0.4.1 ## [53] dbplyr_2.1.1 utf8_1.2.2 ## [55] tidyselect_1.1.2 labeling_0.4.2 ## [57] rlang_1.0.2 later_1.3.0 ## [59] AnnotationDbi_1.59.0 munsell_0.5.0 ## [61] BiocVersion_3.16.0 tools_4.2.0 ## [63] cachem_1.0.6 cli_3.3.0 ## [65] generics_0.1.2 RSQLite_2.2.12 ## [67] ExperimentHub_2.5.0 broom_0.8.0 ## [69] evaluate_0.15 stringr_1.4.0 ## [71] fastmap_1.1.0 yaml_2.3.5 ## [73] knitr_1.38 bit64_4.0.5 ## [75] survMisc_0.5.6 purrr_0.3.4 ## [77] KEGGREST_1.37.0 mime_0.12 ## [79] formatR_1.12 xml2_1.3.3 ## [81] biomaRt_2.53.0 compiler_4.2.0 ## [83] filelock_1.0.2 curl_4.3.2 ## [85] png_0.1-7 interactiveDisplayBase_1.35.0 ## [87] ggsignif_0.6.3 tibble_3.1.6 ## [89] bslib_0.3.1 stringi_1.7.6 ## [91] highr_0.9 GenomicFeatures_1.49.0 ## [93] lattice_0.20-45 KMsurv_0.1-5 ## [95] vctrs_0.4.1 pillar_1.7.0 ## [97] lifecycle_1.0.1 BiocManager_1.30.17 ## [99] jquerylib_0.1.4 data.table_1.14.2 ## [101] bitops_1.0-7 httpuv_1.6.5 ## [103] rtracklayer_1.57.0 R6_2.5.1 ## [105] BiocIO_1.7.0 bookdown_0.26 ## [107] promises_1.2.0.1 gridExtra_2.3 ## [109] codetools_0.2-18 lambda.r_1.2.4 ## [111] assertthat_0.2.1 rjson_0.2.21 ## [113] withr_2.5.0 GenomicAlignments_1.33.0 ## [115] Rsamtools_2.13.0 GenomeInfoDbData_1.2.8 ## [117] hms_1.1.1 tidyr_1.2.0 ## [119] rmarkdown_2.14 carData_3.0-5 ## [121] ggpubr_0.4.0 pROC_1.18.0 ## [123] shiny_1.7.1 restfulr_0.0.13