Load raw counts or normalized sample-level data. Optionally filter by training-regulated features.
For epigenetic data (ATAC and METHYL), if training_regulated_only = FALSE
, sample-level data
is downloaded from Google Cloud Storage.
Usage
load_sample_data(
tissue,
assay,
normalized = TRUE,
training_regulated_only = FALSE,
exclude_outliers = FALSE,
scratchdir = ".",
nrows = Inf,
warnings = TRUE
)
Arguments
- tissue
character, tissue abbreviation, one of MotrpacRatTraining6moData::TISSUE_ABBREV
- assay
character, assay abbreviation, one of MotrpacRatTraining6moData::ASSAY_ABBREV
- normalized
bool, whether to return normalized data. If
FALSE
, return raw counts.- training_regulated_only
bool, whether to filter features down to those training-regulated at 5% FDR
- exclude_outliers
bool, whether to remove sample outliers specified by MotrpacRatTraining6moData::OUTLIERS
- scratchdir
character, local directory in which to download data from Google Cloud Storage. Current working directory by default. Only applies if
assay
is ATAC or METHYL.- nrows
integer, number of rows to return. Defaults to Inf. Useful to return a subset of a large data frame for tests.
- warnings
bool, whether to print warnings to the console.
TRUE
by default.
Value
a data.frame where features are in rows and numeric columns correspond to sample identifiers (vial labels)
Examples
# Load RNA-seq raw counts for liver
data = load_sample_data("LIVER", "TRNSCRPT", normalized = FALSE)
#> TRNSCRPT_LIVER_RAW_COUNTS
# Load normalized metabolomics data for gastrocnemius
data = load_sample_data("SKM-GN", "METAB")
#> METAB SKM-GN normalized data from METAB_NORM_DATA_FLAT
# Load normalized protein abundance data for heart
data = load_sample_data("HEART", "PROT")
#> PROT_HEART_NORM_DATA
if (FALSE) { # \dontrun{
# Load ATAC-seq raw counts for hippocampus, excluding outliers
data = load_sample_data("HIPPOC",
"ATAC",
exclude_outliers = TRUE,
normalized = FALSE,
scratchdir = "/tmp")
} # }