ExtractMotifFromModel: ExtractMotifFromModel allow the user to extract motifs who...

View source: R/ExtractMotifFromModel.R

ExtractMotifFromModelR Documentation

ExtractMotifFromModel allow the user to extract motifs who are activated by the sequences in our model.

Description

ExtractMotifFromModel allow the user to extract motifs who are activated by the sequences in our model.

Usage

ExtractMotifFromModel(
  X = NULL,
  Y = NULL,
  lower.case = F,
  top_kernel = 20,
  model.atac = T
)

Arguments

X

An object of class character,list or DNAStringSet/DNAStringSetList with DNA sequences.

Y

a numeric vector of 1 and 0 values (default to NULL).

lower.case

boolean. Set to TRUE if elements of X are in lower case (default to FALSE).

top_kernel

integer which indicate the top best motifs activated by X.

Details

A kernel is a weight matrix used in convolution layer in DNN (Deep Neural Network). With one-hot DNA, a kernel represented a weighted DNA motif, but can't be easily interpreted. With this function, we detect the location on the response map where the kernel is activated for all input sequences and built the corresponding position count matrix.

Value

A list of Position Count Matrix.


morphos30/DeepG4 documentation built on June 11, 2022, 10:38 p.m.