subset_geo_cv: Create a cross-validation dataset from a known GEO dataset...

In katrinakalantar/metahostlab: Metagenomic Host Labelling of Unlabelled Data (one line, title case)

Description

Create a cross-validation dataset from a known GEO dataset with a covariate of interest.

Usage

subset_geo_cv(geo_dataset_name, geo_dataset_list, pos_regex, neg_regex,
  source_variable, flip_i, flip_j)

Arguments

geo_dataset_name	The dataset name set initially.
geo_dataset_list	The pre-loaded list of GEO datasets.
pos_regex	A regular expression (string) that selects for the positive class in the GEO dataset metadta field specified by source_variable arg
neg_regex	A regular expression (string) that selects for the negative class in the GEO dataset metadta field specified by source_variable arg
source_variable	The GEO dataset variable that will be used to split the dataset.
flip_i	Decimal value indicating the proportion of samples of class B to flip to class A.
flip_j	Decimal value indicating the proportion of samples of class B to flip to class A.

Value

List of datasets split for cross-validation with training and test sets, including flipped labels at specified percentage.

katrinakalantar/metahostlab documentation built on May 26, 2019, 2:31 p.m.