Rf2pval-package: Rf2pval: A comprehensive approach to genomic analysis using...

In tkolisnik/Rf2pval: A comprehensive approach to genomic analysis using scikit-learn's Random Forest models and rank-based feature reduction in R

Rf2pval: A comprehensive approach to genomic analysis using scikit-learn's Random Forest models and rank-based feature reduction in R

Description

Welcome to Rf2pval, a comprehensive tool designed to revolutionize your approach to genomic data analysis with Random Forest Models in R. Whether you're dealing with RNA-seq or Microarray data, Rf2pval integrates scikit-learn's Random Forest methodologies (imported to R via reticulate) for model development, evaluation, and custom feature reduction through rank-based permutation. It also integrates with SHAP and gProfiler for advanced analysis.

Details

The Rf2pval package provides advanced tools for genomic data analysis, featuring seamless integration of scikit-learn's Random Forest methodologies in R. It's designed to be user-friendly for both beginners and advanced users, with a focus on providing accurate and interpretable results. Please see our comprehensive vignette for detailed instructions on using Rf2pval.

Installation

Install Rf2pval from GitHub using devtools:

devtools::install_github("tkolisnik/Rf2pval")

Basic Usage

library(Rf2pval)

Note: This package uses the reticuate R package to use Python's scikit-learn, see vignette for installation and setup instructions. Python 3.9.18 and Conda are required,
Highly recommend to read the vignette and use the demo dataset data("demo_rnaseq_data") for first time users.

reticulate::use_python("/usr/local/bin/python3")
reticulate::conda_create(envname = "rf2pval-conda")
reticulate::conda_install(envname = "rf2pval-conda", packages = c("scikit-learn", "numpy","shap"))
reticulate::use_condaenv("rf2pval-conda", required = TRUE)

data("demo_rnaseq_data")

processed_training_data <- Rf2pval::create_feature_matrix(demo_rnaseq_data$training_data,"training")
processed_validation_data <- Rf2pval::create_feature_matrix(demo_rnaseq_data$training_data,"validation")
processed_testing_data <- Rf2pval::create_feature_matrix(demo_rnaseq_data$training_data,"testing")
tuned_params <- Rf2pval::tune_and_train_rf_model(processed_training_data$X_training_mat, processed_training_data$y_training_vector)
fitting_results<- Rf2pval::fit_and_evaluate_rf(tuning_results$best_params,processed_training_data$X_training_mat,processed_training_data$y_training_vector,processed_validation_data$X_validation_mat,processed_validation_data$y_validation_vector)

... Please see vignette for additional steps and functions, and any issues with setup.

More Information

Please see the comprehensive vignette for detailed instructions: www.github.com/tkolisnik/Rf2pval

Contact and Contribution

We welcome contributions, bug reports, and questions. Please submit issues and pull requests on our GitHub repository or email to [e-mail].

Author(s)

Maintainer: Tyler Kolisnik [e-mail] (ORCID)

tkolisnik/Rf2pval documentation built on Feb. 20, 2024, 5:39 a.m.