% Generated by roxygen2: do not edit by hand

% Please edit documentation in R/predict.R

\name{predict_model}

\alias{predict_model}

\title{Make prediction for nucleotide sequence or entries in fasta/fastq file}

\usage{

predict_model(

  model,

  output_format = "one_seq",

  layer_name = NULL,

  sequence = NULL,

  path_input = NULL,

  round_digits = NULL,

  filename = "states.h5",

  step = 1,

  vocabulary = c("a", "c", "g", "t"),

  batch_size = 256,

  verbose = TRUE,

  return_states = FALSE,

  output_type = "h5",

  padding = "none",

  use_quality = FALSE,

  quality_string = NULL,

  mode = "label",

  lm_format = "target_right",

  output_dir = NULL,

  format = "fasta",

  include_seq = FALSE,

  reverse_complement_encoding = FALSE,

  ambiguous_nuc = "zero",

  ...

)

}

\arguments{

\item{model}{A keras model.}

\item{output_format}{Either \code{"one_seq"}, \code{"by_entry"}, \code{"by_entry_one_file"}, \code{"one_pred_per_entry"}.}

\item{layer_name}{Name of layer to get output from. If \code{NULL}, will use the last layer.}

\item{sequence}{Character string, ignores path_input if argument given.}

\item{path_input}{Path to fasta file.}

\item{round_digits}{Number of decimal places.}

\item{filename}{Filename to store states in. No file output if argument is \code{NULL}.

If \code{output_format = "by_entry"}, adds "\emph{nr}" + "i" after name, where i is entry number.}

\item{step}{Frequency of sampling steps.}

\item{vocabulary}{Vector of allowed characters. Characters outside vocabulary get encoded as specified in \code{ambiguous_nuc}.}

\item{batch_size}{Number of samples used for one network update.}

\item{verbose}{Boolean.}

\item{return_states}{Return predictions as data frame. Only supported for output_format \code{"one_seq"}.}

\item{output_type}{\code{"h5"} or \code{"csv"}. If \verb{output_format`` is }"by_entries_one_file", "one_pred_per_entry"\verb{can only be}"h5"`.}

\item{padding}{Either \code{"none"}, \code{"maxlen"}, \code{"standard"} or \code{"self"}.

\itemize{

\item If \code{"none"}, apply no padding and skip sequences that are too short.

\item If \code{"maxlen"}, pad with maxlen number of zeros vectors.

\item If \code{"standard"}, pad with zero vectors only if sequence is shorter than maxlen. Pads to minimum size required for one prediction.

\item If \code{"self"}, concatenate sequence with itself until sequence is long enough for one prediction.

Example: if sequence is "ACGT" and maxlen is 10, make prediction for "ACGTACGTAC".

Only applied if sequence is shorter than maxlen.

}}

\item{use_quality}{Whether to use quality scores.}

\item{quality_string}{String for encoding with quality scores (as used in fastq format).}

\item{mode}{Either \code{"lm"} for language model or \code{"label"} for label classification.}

\item{lm_format}{Either \code{"target_right"}, \code{"target_middle_lstm"}, \code{"target_middle_cnn"} or \code{"wavenet"}.}

\item{output_dir}{Directory for file output.}

\item{format}{File format, \code{"fasta"}, \code{"fastq"}, \code{"rds"} or \code{"fasta.tar.gz"}, \code{"fastq.tar.gz"} for \code{tar.gz} files.}

\item{include_seq}{Whether to include input sequence in h5 file.}

\item{reverse_complement_encoding}{Whether to use both original sequence and reverse complement as two input sequences.}

\item{ambiguous_nuc}{How to handle nucleotides outside vocabulary, either \code{"zero"}, \code{"discard"}, \code{"empirical"} or \code{"equal"}.

\itemize{

\item If \code{"zero"}, input gets encoded as zero vector.

\item If \code{"equal"}, input is repetition of \code{1/length(vocabulary)}.

\item If \code{"discard"}, samples containing nucleotides outside vocabulary get discarded.

\item If \code{"empirical"}, use nucleotide distribution of current file.

}}

\item{...}{Further arguments for sequence encoding with \code{\link{seq_encoding_label}}.}

}

\value{

If \code{return_states = TRUE} returns a list of model predictions and position of corresponding sequences.

If additionally \code{include_seq = TRUE}, list contains sequence strings.

If \code{return_states = FALSE} returns nothing, just writes output to file(s).

}

\description{

Removes layers (optional) from pretrained model and calculates states of fasta/fastq file or nucleotide sequence.

Writes states to h5 or csv file (access content of h5 output with \code{\link{load_prediction}} function).

There are several options on how to process an input file:

\itemize{

\item If \code{"one_seq"}, computes prediction for sequence argument or fasta/fastq file.

Combines fasta entries in file to one sequence. This means predictor sequences can contain elements from more than one fasta entry.

\item If \code{"by_entry"}, will output a separate file for each fasta/fastq entry.

Names of output files are: \code{output_dir} + "Nr" + i + \code{filename} + \code{output_type}, where i is the number of the fasta entry.

\item If \code{"by_entry_one_file"}, will store prediction for all fasta entries in one h5 file.

\item If \code{"one_pred_per_entry"}, will make one prediction for each entry by either picking random sample for long sequences

or pad sequence for short sequences.

}

}

\examples{

\dontshow{if (reticulate::py_module_available("tensorflow")) (if (getRversion() >= "3.4") withAutoprint else force)(\{ # examplesIf}

# make prediction for single sequence and write to h5 file

model <- create_model_lstm_cnn(maxlen = 20, layer_lstm = 8, layer_dense = 2, verbose = FALSE)

vocabulary <- c("a", "c", "g", "t")

sequence <- paste(sample(vocabulary, 200, replace = TRUE), collapse = "")

output_file <- tempfile(fileext = ".h5")

predict_model(output_format = "one_seq", model = model, step = 10,

             sequence = sequence, filename = output_file, mode = "label")

# make prediction for fasta file with multiple entries, write output to separate h5 files

fasta_path <- tempfile(fileext = ".fasta")

create_dummy_data(file_path = fasta_path, num_files = 1,

                 num_seq = 5, seq_length = 100,

                 write_to_file_path = TRUE)

model <- create_model_lstm_cnn(maxlen = 20, layer_lstm = 8, layer_dense = 2, verbose = FALSE)

output_dir <- tempfile()

dir.create(output_dir)

predict_model(output_format = "by_entry", model = model, step = 10, verbose = FALSE,

               output_dir = output_dir, mode = "label", path_input = fasta_path)

list.files(output_dir)

\dontshow{\}) # examplesIf}

}