size.effect {permubiome}

Executing estimation statistics based on bootstrap-coupled approach

Description

Assessing the size effect on selected microbiome features found to be differentially abundant between classes. This analysis is based on the Data Analysis using Bootstrap-Coupled Estimation (dabestr) R package and gives you the option to create Gardner-Altman estimation plots individually all features found to be differentially presented in your dataset.

Usage

size.effect(category = "", replicates = 5000, 
paired = FALSE, plot.file = "tiff", id.pairs = NULL)

Arguments

category

Name of the microbiome feature, which differential abundance between classes will be further explored.

replicates

The number of bootstrap resamples that have to be generated. Integer, default 5000.

paired

If TRUE, the two groups are treated as paired samples, please add an extra column (id.pairs) to parse identity of the datapoint. Default FALSE, the control_group group is treated as pre-intervention and the test_group group is considered post-intervention.

plot.file

Extension for plot graphics (ps, pdf, jpeg, tiff, png, bmp). Default "tiff".

id.pairs

Column name for information to parse identity of the datapoint in case of paired data.

Details

Be careful to type the "category" correctly to be analyzed in order to that matches with the table contained information.

Author(s)

Alfonso Benitez-Paez

References

Benitez-Paez A. 2023. Permubiome: an R package to perform permutation based test for biomarker discovery in microbiome analyses. [https://cran.r-project.org]. Benitez-Paez A, et al. mSystems. 2020;5:e00857-19. doi: 10.1128/mSystems.00857-19.

Examples

## The function is currently defined as
function (category = "", replicates = 5000, 
    paired = FALSE, plot.file = "tiff", id.pairs = NULL)
{
    Class <- NULL
    ref <- NULL
    loadNamespace("dabestr")
    loadNamespace("rlang")
    loadNamespace("dplyr")
    load("permubiome.RData")
    df_norm <- df_norm
    if (paired == TRUE) {
        print(paste("You declared paired data, be sure to include the correct -id.column- argument 
	to parse the identity of the datapoint!"))
    }
    classes <- levels(df_norm$Class)
    if (REFERENCE == "") {
        REFERENCE <- classes[1]
    }
    else if (REFERENCE == classes[2]) {
        classes[2] <- classes[1]
        classes[1] <- REFERENCE
    }
    df_norm<-tibble(df_norm)
    prepare.stats <- load(df_norm, Class, category, paired = paired, 
        idx = c(classes[1], classes[2]), id_col = id.pairs)
    prepare.stats$y<-quo_set_expr(prepare.stats$y, as.symbol(category))
    print(prepare.stats)
    if (category == "") {
        category <- colnames(df_norm[3])
        print(paste("As you declared no categories, the very first one of your dataset will be 
	processed!"))
    }
    estimation.stats<-median_diff(prepare.stats, perm_count = replicates)
    e_plot <- plot(estimation.stats, group.summaries = "median_quartiles", 
        palette = "Set1", rawplot.ylabel = paste(category, "normalized reads", 
            sep = " "), tick.fontsize = 12, axes.title.fontsize = 18)
    tiff(filename = paste(category, "estimation", plot.file, sep = "."), 
        width = 650, height = 600, res = 100, units = "px")
    e_plot
    dev.off()
    print(e_plot)
    save(df, df_norm, REFERENCE, classes, file = "permubiome.RData")
  }