In ruthkr/deepredeff: Deep Learning Prediction of Effectors

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-",
  out.width = "100%"
)

ggplot2::theme_set(ggplot2::theme_bw())

deepredeff

deepredeff is a package to predict effector protein given amino acid sequences. This tool can be used to predict effectors from three different taxa, which are oomycete, fungi, and bacteria.

Installation

You can install the released version of deepredeff from CRAN with:

install.packages("deepredeff")

And the development version from GitHub with:

# install.packages("devtools")
devtools::install_github("ruthkr/deepredeff")

The deepredeff package uses TensorFlow. If you already have TensorFlow 2.0.0 or later in your system, then you can specify the environment where TensorFlow is installed using reticulate::use_condaenv(). Otherwise, you can install TensorFlow, by using the install_tensorflow() function as follows:

library(deepredeff)
install_tensorflow()

Note that this only needs to be run once, the first time you use deepredeff.

Documentation

To use deepredeff, you can read the documentation on the following topics:

1. Getting started
2. Effector prediction with various different input formats and models

Quick start

This is a basic example which shows you how to predict effector sequences if you have a FASTA file:

# Load the package
library(deepredeff)

# Define the fasta path from the sample data
bacteria_fasta_path <- system.file(
  "extdata/example", "bacteria_sample.fasta", 
  package = "deepredeff"
)

# Predict the effector candidate using bacteria model
pred_result <- predict_effector(
  input = bacteria_fasta_path,
  taxon = "bacteria"
)

# View results
pred_result

pred_result %>%
  dplyr::mutate(
    name = stringr::str_replace_all(name, "\\|", "⎮"),
    sequence = stringr::str_sub(sequence, 1, 25)
  ) %>%
  knitr::kable()

After getting the prediction results, you can plot the probability distribution of the results as follows:

plot(pred_result)

More examples with different input formats are available on functions documentations and vignettes, please refer to the documentation.

ruthkr/deepredeff documentation built on Sept. 18, 2023, 4:25 a.m.