prepModelData: Prepare Dataset for Modeling
In PriceLab/FPML: Footprint Machine Learning

Prepare the fully annotated FIMO motif dataset for running models by filtering it accordingly

1 2	prepModelData(annotated.df, seed, hintCutoff = -Inf, wellCutoff = Inf, motifsOnly = FALSE, biasTraining = FALSE, biasRatio = 9)

`annotated.df`	The fully annotated FIMO motif dataset. This must be a data frame, or else the function will stop and give an error.
`seed`	A character string that should be one of c("16","20","both"). This seed should match the data in the data frame
`hintCutoff`	A cutoff value used as a threshold on the HINT footprint score. Only scores above this threshold will be kept (default = -Inf)
`wellCutoff`	A cutoff value used as a threshold on the Wellington footprint score. Only scores below this threshold will be kept (default = Inf)
`motifsOnly`	A Boolean value indicating whether to prepare data such that it uses only motifs, neglecting the footprint data (default = FALSE)

A list in which the filtered data frame has been split into 6 pieces for X/Y and training/testing/validation sets. These sets are completely ready to run gradient boosted and linear models and should be passed as a list to those commands accordingly

PriceLab/FPML documentation built on May 28, 2019, 2:25 p.m.