README.md

In sahilseth/cghub: An Interface enabling easy access to CGHub

cghub: An interface for chatting with Cancer Genomics Hub

April 7, 2015

cghub.ucsc.edu

this package is under active developement Feel free to fork, log issues and submit pull requests https://github.com/sahilseth/cghub

Installing the package

install.packages("devtools")
devtools::install_github("sahilseth/cghub")

Examples:

Querying

For more on querying the database, please look at: https://cghub.ucsc.edu/docs/user/query.html

Get hits for a tumor type

Query is based on: - Get all WGS bams which are live for tumor CESC, and provide the number available

library(cghub)
out <- gt.cgquery(search = "tumor=CESC&library_strategy=WGS&study=phs000178&state=live",
                  count.only = TRUE)

Query and parse hits into a table

s = c(
  study = 'phs000178',
  library_strategy = 'WGS',
  disease_abbr = 'CESC',
  state = 'live')
search = paste(names(s), s, collapse = "&", sep = "=")
#search
out <- gt.cgquery(search = search, count.only = FALSE)
names(out)

[1] "tab" "query" "hits"

out$query

[1] "study:phs000178 library_strategy:WGS disease_abbr:CESC state:live"

out$hits

[1] 142

|study |disease |platform |centerName |barcode |aliquotID |reflib |filename |filesize |state |uuid |checksum |modificationDt |reason |libraryStrategy | |:---------|:-------|:--------|:----------|:----------------------------|:------------------------------------|:------------------|:-------------------------------------------------------------------------|:-----------|:-----|:------------------------------------|:--------------------------------|:--------------------|:------|:---------------| |phs000178 |CESC |ILLUMINA |HMS-RK |TCGA-HG-A2PA-01A-11D-A20X-26 |ECE06317-01F9-4EF2-ACE0-E9B18C7F4DE1 |HG19_Broad_variant |TCGA-HG-A2PA-01A-11D-A20X_140106_SN590_0244_BC384EACXX_s_4_rg.sorted.bam |17463115455 |live |03ad648b-e0e2-4d73-a1a7-1004a4cf6c6d |52fbc3079d348d0f3e0b400e7be185e2 |2014-06-10T05:21:37Z |NA |WGS | |phs000178 |CESC |ILLUMINA |HMS-RK |TCGA-FU-A3TQ-01A-11D-A232-26 |7018744f-49ff-4f47-aee5-07f0dea5d7d3 |HG19_Broad_variant |TCGA-FU-A3TQ-01A-11D-A232_140124_SN1120_0291_AC2PHAACXX_s_3_rg.sorted.bam |26168324162 |live |03f34fcc-2af0-41f5-9784-3de867b1aac3 |f951fb705fc93a9848d630df633c0580 |2014-06-10T07:16:47Z |NA |WGS | |phs000178 |CESC |ILLUMINA |HMS-RK |TCGA-JX-A3Q0-10A-01D-A245-26 |3494d848-1830-49a0-b144-22550b8a0428 |HG19_Broad_variant |TCGA-JX-A3Q0-10A-01D-A245_140206_SN1222_0239_BC2RAKACXX_s_2_rg.sorted.bam |19341942149 |live |10e36cd3-f435-4e63-83f9-57613bb50b90 |7a0905a27831dff3537a6ce443397a00 |2014-06-10T18:50:50Z |NA |WGS | |phs000178 |CESC |ILLUMINA |HMS-RK |TCGA-JX-A3Q8-10A-01D-A245-26 |dfede0ba-0e85-4d22-9a70-88e798cd267a |HG19_Broad_variant |TCGA-JX-A3Q8-10A-01D-A245_140206_SN1222_0239_BC2RAKACXX_s_4_rg.sorted.bam |20895427729 |live |0fbd6804-e8d6-47b5-95f7-aa3f107d2c7e |43e830a660e342ce7251032caa436ff8 |2014-06-10T18:57:02Z |NA |WGS | |phs000178 |CESC |ILLUMINA |HMS-RK |TCGA-C5-A2LZ-01A-11D-A20X-26 |837B293C-0843-4002-A8AB-3A945E040C75 |HG19_Broad_variant |TCGA-C5-A2LZ-01A-11D-A20X_140106_SN590_0244_BC384EACXX_s_2_rg.sorted.bam |16607119021 |live |0a7b3c97-5fcf-493b-811f-2eb9734bae1a |ba54f79d49539a64373e831b9a98804a |2014-06-11T01:53:18Z |NA |WGS | |phs000178 |CESC |ILLUMINA |HMS-RK |TCGA-EA-A3HU-10B-01D-A20X-26 |4B33B319-40C6-4C3C-B259-5518A1B55410 |HG19_Broad_variant |TCGA-EA-A3HU-10B-01D-A20X_140106_SN590_0244_BC384EACXX_s_7_rg.sorted.bam |18612972742 |live |03dd6beb-53b9-4bfe-8843-9e2fe621913a |5158447a4b01a7e5ef91bafa3f2a81c2 |2014-06-12T00:59:54Z |NA |WGS |

Based on a list of TCGA barcodes

require(parallel)
ids <- c("TCGA-BJ-A0Z5-01","TCGA-DJ-A1QG-01","TCGA-EM-A1YC-01",
         "TCGA-ET-A25J-01", "TCGA-H2-A26U-01","TCGA-H2-A3RI-01")

tab <- mclapply(ids,function(id){
  out <- gt.cgquery(search=sprintf("legacy_sample_id=%s*&state=live",id), count.only=FALSE)
  return(out$tab)
  },mc.cores=1)
tab=do.call(rbind,tab)
knitr::kable(head(out$tab), format = "markdown")

Fetching the data

Load BAM using GTFuse

## This contains all the uuids: out$tab$uuid
path=gt.loadBamFile(uuid='056a5a41-1be5-48f1-8ae5-bd8125e5b0c2', key="mykey.pem", force=TRUE)

Download BAM using gtdownload

• this supports providing multiple UUIDs

outBam <- gt.cgDownload(uuids=uuids, path="/scratch/gt", 
                        GT="gtdownload", cores=1,
                        key="mykey.pem",params="")

sahilseth/cghub documentation built on May 29, 2019, 12:57 p.m.