fastaFileGenerator: Custom generator for fasta/fastq files

In hiddengenome/altum: deepG

Description

fastaFileGenerator Iterates over folder containing .fasta/.fastq files and produces one-hot-encoding of predictor sequences and target variables.

Usage

fastaFileGenerator(
  corpus.dir,
  format = "fasta",
  batch.size = 256,
  maxlen = 250,
  max_iter = 2000,
  vocabulary = c("a", "c", "g", "t"),
  verbose = FALSE,
  randomFiles = FALSE,
  step = 1,
  showWarnings = FALSE,
  seed = 1234,
  shuffleFastaEntries = FALSE,
  numberOfFiles = NULL,
  fileLog = NULL,
  reverseComplements = FALSE
)

Arguments

corpus.dir	Input directory where .fasta files are located or path to single file ending with .fasta or .fastq (as specified in format argument).
format	File format, either fasta or fastq.
batch.size	Number of batches.
maxlen	Length of predictor sequence.
max_iter	Stop after max_iter number of iterations failed to produce a new batch.
vocabulary	Vector of allowed characters, character outside vocabulary get encoded as 0-vector.
verbose	Whether to show message.
randomFiles	Logical, whether to go through files randomly or sequential.
step	How often to take a sample.
showWarnings	Logical, give warning if character outside vocabulary appears
seed	Sets seed for set.seed function, for reproducible results when using randomFiles or shuffleFastaEntries
shuffleFastaEntries	Logical, shuffle entries in every fasta file before connecting them to sequence.
numberOfFiles	Use only specified number of files, ignored if greater than number of files in corpus.dir.
fileLog	Write name of files to csv file if path is specified.
reverseComplements	Logical, half of batch contains sequences and other its reverse complements. Reverse complement is given by reversed order of sequence and switching A/T and C/G. batch.size argument has to be even, otherwise 1 will be added to batch.size

Value

A list of length 2. First element is a 3-dimensional tensor with dimensions (batch.size, maxlen, length(vocabulary)), encoding the predictor sequences. Second element is a matrix with dimensions (batch.size, length(vocabulary)), encoding the targets.

hiddengenome/altum documentation built on April 22, 2020, 9:33 p.m.