Entering edit mode
cschot3
•
0
@cschot3-18570
Last seen 6.0 years ago
I recently updated R and some packages and ran into a problem when trying to run GEOquery to download a dataset. When grabbing the expression set from the dataset, I noticed that the sample names were no longer the column names, but that expression values were now the column names, almost as if the first row was taken from the dataset and used as the header. Is this a bug with GEOquery, Biobase, or NCBI's database? Any solutions?
$`GSE38958_series_matrix.txt.gz` ExpressionSet (storageMode: lockedEnvironment) assayData: 21787 features, 115 samples element names: exprs protocolData: none phenoData sampleNames: 6.156901 6.317044 ... 6.314153 (115 total) varLabels: title geo_accession ... data_row_count (35 total) varMetadata: labelDescription featureData featureNames: 2315633 2315674 ... 7385696 (21787 total) fvarLabels: ID GB_LIST ... category (12 total) fvarMetadata: Column Description labelDescription experimentData: use 'experimentData(object)' Annotation: GPL5175 > sessionInfo() R version 3.5.1 (2018-07-02) Platform: x86_64-w64-mingw32/x64 (64-bit) Running under: Windows >= 8 x64 (build 9200) Matrix products: default locale: [1] LC_COLLATE=English_United States.1252 LC_CTYPE=English_United States.1252 [3] LC_MONETARY=English_United States.1252 LC_NUMERIC=C [5] LC_TIME=English_United States.1252 attached base packages: [1] parallel stats graphics grDevices utils datasets methods base other attached packages: [1] bindrcpp_0.2.2 GEOquery_2.48.0 Biobase_2.40.0 BiocGenerics_0.26.0 loaded via a namespace (and not attached): [1] Rcpp_1.0.0 rstudioapi_0.8 bindr_0.1.1 xml2_1.2.0 magrittr_1.5 [6] hms_0.4.2 tidyselect_0.2.5 R6_2.3.0 rlang_0.3.0.1 dplyr_0.7.8 [11] tools_3.5.1 assertthat_0.2.0 tibble_1.4.2 crayon_1.3.4 BiocManager_1.30.4 [16] purrr_0.2.5 readr_1.2.1 tidyr_0.8.2 curl_3.2 glue_1.3.0 [21] limma_3.36.5 stringi_1.2.4 compiler_3.5.1 pillar_1.3.0 pkgconfig_2.0.2
This is a bug due to a change in behavior of the readr read_tsv function. https://github.com/tidyverse/readr/issues/925. There is an ongoing conversation about what to do here. Sorry for the instability.
Just a quick update--the Rstudio team has a working fix and it is going through code review. Once that is completed and the new readr package is released, the GEOquery problem will be resolved.