pancan_SE: illustrate DelayedArray assay from BigQuery backend in...

In restfulSE: Access matrix-like HDF5 server content or BigQuery content through a SummarizedExperiment interface

Description

illustrate DelayedArray assay from BigQuery backend in SummarizedExperiment

Usage

pancan_SE(
  bqcon,
  colDataTableName = "clinical_PANCAN_patient_with_followup",
  clinVars = basic_clinvars(),
  colDSubjectIdName = "bcr_patient_barcode",
  colDFilterField = "acronym",
  colDFilterValue = "BRCA",
 
    assayDataTableName = "pancanMiRs_EBadjOnProtocolPlatformWithoutRepsWithUnCorrectMiRs_08_04_16_annot",
  assayFeatureName = "ID",
  assaySampleTypeCode = "TP",
  subjectIDName = "ParticipantBarcode",
  tumorFieldName = "Study",
  tumorFieldValue = "BRCA",
  assayValueFieldName = "miRNAexpr"
)

Arguments

bqcon	a BigQueryConnection instance
colDataTableName	character(1) defaulting to "clinical_PANCAN_patient_with_followup"
clinVars	character() vector of variables to be retained from the table named by 'colDataTableName', defaults to vector returned by clinVars()
colDSubjectIdName	character(1) defaulting to "bcr_patient_barcode"
colDFilterField	character(1) defaulting to "acronym"
colDFilterValue	character(1) defaulting to "BRCA"; a vector may be used, in which case multiple tumor types will be represented – must agree with tumorFieldValue, as clinical and assay data are collected separately
assayDataTableName	character(1) defaulting to "pancanMiRs_EBadjOnProtocolPlatformWithoutRepsWithUnCorrectMiRs_08_04_16_annot"
assayFeatureName	character(1) defaulting to "ID"
assaySampleTypeCode	character(1) defaulting to "TP"
subjectIDName	character(1) defaulting to "ParticipantBarcode"
tumorFieldName	character(1) defaulting to "Study"
tumorFieldValue	character() defaulting to "BRCA"; a vector may be used in which case multiple tumor types will be represented – must agree with colDFilterValue
assayValueFieldName	character(1) defaulting to "miRNAexpr"

Value

an instance of SummarizedExperiment

Note

The parameters need different assignments for different tables. Field names are not standardized across tables as of August 2018. AUTHENTICATION CONCERNS: You must have a valid BigQuery project identifier in the environment variable CGC_BILLING, or pass such as 'billing' when using DBI::dbConnect(bigquery::bigrquery(), ...). To get such a project identifier, you need to have a Google identity and you must have created a BigQuery project with that identity. Notes at https://isb-cancer-genomics-cloud.readthedocs.io/en/latest/sections/progapi/bigqueryGUI/WalkthroughOfGoogleBigQuery.html provide details. Another approach that can be used involves the Google Cloud SDK. Assuming you have a Google identity and have installed a recent SDK instance, you can use (in the shell) gcloud auth login to pick the identity that has a project with id PROJECT_ID. Use that project id as the billing code for bigrquery dbConnect, or set CGC_BILLING in the environment to evaluate to that project id.

Examples

if (interactive() & requireNamespace("DBI") & 
   requireNamespace("bigrquery")) {
     billco = Sys.getenv("CGC_BILLING")
     if (nchar(billco)==0) 
          stop("need CGC_BILLING set to your BigQuery project ID, see note in ?pancan_SE")
     bqcon = DBI::dbConnect(bigrquery::bigquery(), project = "pancancer-atlas", 
           dataset = "Annotated", billing = billco)
     brca_mirSE = pancan_SE(bqcon)
     brca_mirSE
     }

restfulSE documentation built on Nov. 8, 2020, 8:01 p.m.