UCSCXenaTools介绍

生信技能树

发布于 2019-10-24 19:27:58

1.9K0

发布于 2019-10-24 19:27:58

UCSCXenaTools 提供了下载 UCSC Xena 平台数据的 R 客户端，为官方文档 https://ucscxena.gitbook.io/project/overview-of-features/download-data 所推荐。

安装

从 CRAN 安装稳定版本：

install.packages("UCSCXenaTools")

从 GitHub 安装开发版本：

# install.packages("remotes")
remotes::install_github("ropensci/UCSCXenaTools")

如果你想要确保在本地构建好包文档，请添加额外两个选项：

remotes::install_github("ropensci/UCSCXenaTools", build_vignettes = TRUE, dependencies = TRUE)

如果存在问题或是 Bug，请到 GitHub Issue 上提问。

Data Hub 列表

UCSC Xena 平台根据数据源分为不同的 data hub，（基本上）所有的数据集都可以在 https://xenabrowser.net/datapages/ 找到。

当前 UCSCXenaTools 支持 10 个 data hub：

UCSC Public Hub: https://ucscpublic.xenahubs.net
TCGA Hub: https://tcga.xenahubs.net
GDC Xena Hub: https://gdc.xenahubs.net
ICGC Xena Hub: https://icgc.xenahubs.net
Pan-Cancer Atlas Hub: https://pancanatlas.xenahubs.net
GA4GH (TOIL) Hub: https://toil.xenahubs.net
Treehouse Hub: https://xena.treehouse.gi.ucsc.edu
PCAWG Hub: https://pcawg.xenahubs.net
ATAC-seq Hub: https://atacseq.xenahubs.net
Singel Cell Xena hub: https://singlecell.xenahubs.net

如果 data hub 的 URL 改变了，或者有新的 data hub 出现了，请通过邮箱 w_shixiang@163.com 或者 GitHub issue 联系我。

使用方法

我将该包支持的标准流程分为 5 步，它们分别由对应的函数实现：

生成 XenaHub 对象 —— XenaGenerate()
过滤数据 —— XenaFilter()
检索数据 —— XenaQuery()
下载数据 —— XenaDownload()
导入数据 —— XenaPrepare()

它们可以使用管道符号 %>% 进行连接。

下面通过下载 TCGA hub 的肺癌临床数据进行演示。

XenaData 数据框

UCSCXenaTools 使用包内置数据集 XenaData 辅助生成 XenaHub 对象，这个数据集记录了当前所有数据集的信息。

library(UCSCXenaTools)
data(XenaData)

head(XenaData)
#> # A tibble: 6 x 17
#>   XenaHosts XenaHostNames XenaCohorts XenaDatasets SampleCount DataSubtype
#>   <chr>     <chr>         <chr>       <chr>              <int> <chr>      
#> 1 https://… publicHub     Breast Can… ucsfNeve_pu…          51 gene expre…
#> 2 https://… publicHub     Breast Can… ucsfNeve_pu…          57 phenotype  
#> 3 https://… publicHub     Glioma (Ko… kotliarov20…         194 copy number
#> 4 https://… publicHub     Glioma (Ko… kotliarov20…         194 phenotype  
#> 5 https://… publicHub     Lung Cance… weir2007_pu…         383 copy number
#> 6 https://… publicHub     Lung Cance… weir2007_pu…         383 phenotype  
#> # … with 11 more variables: Label <chr>, Type <chr>,
#> #   AnatomicalOrigin <chr>, SampleType <chr>, Tags <chr>, ProbeMap <chr>,
#> #   LongTitle <chr>, Citation <chr>, Version <chr>, Unit <chr>,
#> #   Platform <chr>

Workflow

生成对象并过滤数据集：

# The options in XenaFilter function support Regular Expression
XenaGenerate(subset = XenaHostNames=="tcgaHub") %>% 
  XenaFilter(filterDatasets = "clinical") %>% 
  XenaFilter(filterDatasets = "LUAD|LUSC|LUNG") -> df_todo

df_todo
#> class: XenaHub 
#> hosts():
#>   https://tcga.xenahubs.net
#> cohorts() (3 total):
#>   TCGA Lung Cancer (LUNG)
#>   TCGA Lung Adenocarcinoma (LUAD)
#>   TCGA Lung Squamous Cell Carcinoma (LUSC)
#> datasets() (3 total):
#>   TCGA.LUNG.sampleMap/LUNG_clinicalMatrix
#>   TCGA.LUAD.sampleMap/LUAD_clinicalMatrix
#>   TCGA.LUSC.sampleMap/LUSC_clinicalMatrix

有时候我们仅仅知道一些关键字词， XenaScan() 函数可以用于逐行扫描 XenaData 所有的列。

x1 = XenaScan(pattern = 'Blood')
x2 = XenaScan(pattern = 'LUNG', ignore.case = FALSE)

x1 %>%
    XenaGenerate()
#> class: XenaHub 
#> hosts():
#>   https://ucscpublic.xenahubs.net
#>   https://tcga.xenahubs.net
#> cohorts() (6 total):
#>   Connectivity Map
#>   TARGET Acute Lymphoblastic Leukemia
#>   Pediatric tumor (Khan)
#>   Acute lymphoblastic leukemia (Mullighan 2008)
#>   TCGA Pan-Cancer (PANCAN)
#>   TCGA Acute Myeloid Leukemia (LAML)
#> datasets() (34 total):
#>   cmap/rankMatrix_reverse
#>   TARGET_ALL/TARGETcnv_genomicMatrix
#>   TARGET_ALL/TARGETexp_genomicMatrix
#>   ...
#>   TCGA.LAML.sampleMap/mutation_wustl
#>   TCGA.LAML.sampleMap/Pathway_Paradigm_RNASeq_And_Copy_Number
x2 %>%
    XenaGenerate()
#> class: XenaHub 
#> hosts():
#>   https://tcga.xenahubs.net
#> cohorts() (1 total):
#>   TCGA Lung Cancer (LUNG)
#> datasets() (13 total):
#>   TCGA.LUNG.sampleMap/HumanMethylation27
#>   TCGA.LUNG.sampleMap/HumanMethylation450
#>   TCGA.LUNG.sampleMap/Gistic2_CopyNumber_Gistic2_all_data_by_genes
#>   ...
#>   TCGA.LUNG.sampleMap/HiSeqV2_exon
#>   TCGA.LUNG.sampleMap/AgilentG4502A_07_3

检索和下载：

XenaQuery(df_todo) %>%
  XenaDownload() -> xe_download
#> This will check url status, please be patient.
#> All downloaded files will under directory /var/folders/mx/rfkl27z90c96wbmn3_kjk8c80000gn/T//RtmpNjqKIQ.
#> The 'trans_slash' option is FALSE, keep same directory structure as Xena.
#> Creating directories for datasets...
#> Downloading TCGA.LUNG.sampleMap/LUNG_clinicalMatrix.gz
#> Downloading TCGA.LUAD.sampleMap/LUAD_clinicalMatrix.gz
#> Downloading TCGA.LUSC.sampleMap/LUSC_clinicalMatrix.gz

导入 R ：

cli = XenaPrepare(xe_download)
class(cli)
#> [1] "list"
names(cli)
#> [1] "LUNG_clinicalMatrix.gz" "LUAD_clinicalMatrix.gz"
#> [3] "LUSC_clinicalMatrix.gz"

浏览数据集

创建两个 XenaHub 对象

to_browse - 包含 1 个队列 1 个数据集
to_browse2 - 包含 2 个队列 2 个数据集

XenaGenerate(subset = XenaHostNames=="tcgaHub") %>%
    XenaFilter(filterDatasets = "clinical") %>%
    XenaFilter(filterDatasets = "LUAD") -> to_browse

to_browse
#> class: XenaHub 
#> hosts():
#>   https://tcga.xenahubs.net
#> cohorts() (1 total):
#>   TCGA Lung Adenocarcinoma (LUAD)
#> datasets() (1 total):
#>   TCGA.LUAD.sampleMap/LUAD_clinicalMatrix

XenaGenerate(subset = XenaHostNames=="tcgaHub") %>%
    XenaFilter(filterDatasets = "clinical") %>%
    XenaFilter(filterDatasets = "LUAD|LUSC") -> to_browse2

to_browse2
#> class: XenaHub 
#> hosts():
#>   https://tcga.xenahubs.net
#> cohorts() (2 total):
#>   TCGA Lung Adenocarcinoma (LUAD)
#>   TCGA Lung Squamous Cell Carcinoma (LUSC)
#> datasets() (2 total):
#>   TCGA.LUAD.sampleMap/LUAD_clinicalMatrix
#>   TCGA.LUSC.sampleMap/LUSC_clinicalMatrix

XenaBrowse() 可以用于浏览数据所在的 UCSC Xena 页面，使用默认浏览器打开。默认情况下，我只允许用户一次打开一个网页，以避免页面过多。

# 运行下面会打开浏览器
XenaBrowse(to_browse)

XenaBrowse(to_browse, type = "cohort")
# 运行下面会报错
XenaBrowse(to_browse2)
#> Error in XenaBrowse(to_browse2): This function limite 1 dataset to browse.
#>  Set multiple to TRUE if you want to browse multiple links.

XenaBrowse(to_browse2, type = "cohort")
#> Error in XenaBrowse(to_browse2, type = "cohort"): This function limite 1 cohort to browse. 
#>  Set multiple to TRUE if you want to browse multiple links.

如果你确定你想要浏览多个页面，也可以通过设定实现：

XenaBrowse(to_browse2, multiple = TRUE)
XenaBrowse(to_browse2, type = "cohort", multiple = TRUE)