Generate 'BART' File

A set of functions to generate and format results from statistical analyses of a wide range of high throughput experiments that can then be uploaded into the 'BART' (Bio-Statistical Analysis Reporting Tool) 'shiny' app < https://github.com/jcardenas14/BART>. The app provides users with tools to visualize and efficiently sift through large amounts of data and results.


Build Status

genBart

The goal of `genBart` is to streamline the process of statistical analysis in high dimensional studies (e.g. RNA-seq and microarray) to a final and comprehensive results object that can be uploaded into the BART (Biostatistical Analysis Reporting Tool) shiny app. The app provides users with tools to interactively visualize and efficiently sift through large amounts of data and results.

The genBart package accomplishes its goal by providing a set of functions that modularize every step of the analysis workflow:

  1. Summary stats and quality control
  2. Unsupervised analysis (heat maps and cluster analysis)
  3. DGE (differential gene expression)
  4. Gene set analysis
  5. Correlation analysis

Though one strength of BART is its ability to report the entire analysis workflow in one session, it is often useful to use BART before all of the analysis is completed (e.g. viewing heat maps before any statistical analysis is run). genBart makes this possible by allowing users to easily update and/or add to existing BART result objects. Conveniently, BART will only populate with tools based on the information contained in the object generated by genBart.

Installation

You can install genBART from github with:

# install.packages("devtools")
devtools::install_github("jcardenas14/genBart")

Example

I demonstrate a simple example below of how BART can be used for unsupervised analysis. The data used for this example is available in genBart and is taken from a longitudinal microarray experiment monitoring the gene expression changes in cynomolgus macaques infected with M.tuberculosis (Skinner et al.). To speed up the hierarchical clustering step, I randomly selected a subset (1000) of the probes for the example. Please see the genBart vignette for a full analysis workflow walk-through from the same microarray study.

## Call data
library(genBart)
data(tb.expr)
data(tb.design)
 
# declare design information
meta <- metaData(y = tb.expr, design = tb.design, data.type = "microarray", 
                 columnname = "columnname", long = TRUE, 
                 subject.id = "monkey_id", baseline.var = "timepoint", 
                 baseline.val = 0, time.var = "timepoint", sample.id = "sample_group")
                    
# normalize data and cluster  
norm.data <- normalizeData(meta = meta, norm.method = "mean")
dendros <- clusterData(norm.data = norm.data)
 
# create BART result object
genFile(meta = list(meta), dendrograms = dendros, project.name = "TB Unsupervised Analysis")

News

genBart 1.0.0

  • First public release

genBart 1.0.1

  • Patched version. genBart 1.0.0 was archived due to archiving of NMF package, which is now back on CRAN.

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.

install.packages("genBart")

1.0.1 by Jacob Cardenas, 9 months ago


Browse source code at https://github.com/cran/genBart


Authors: Jacob Cardenas [aut, cre] , Jacob Turner [aut] , Derek Blankenship [aut]


Documentation:   PDF Manual  


GPL-2 | GPL-3 license


Imports clValid, data.table, fastcluster, ggplot2, grid, gtools, limma, NMF, pca3d, psych, qusage, RColorBrewer, reshape2, rmarkdown, scales, shiny, shinydashboard, shinyjs, statmod, stats, stringr, tools, VennDiagram

Suggests DESeq2, edgeR, knitr, SummarizedExperiment, testthat


See at CRAN