R/vsMAPlot.R

In btmonier/vidger: Create rapid visualizations of RNAseq data in R

#' @title
#' MA plot from mean expression and \eqn{log} fold changes from different
#' analytical objects
#'
#' @author
#' Brandon Monier, \email{brandon.monier@sdstate.edu}
#'
#' @description
#' This function allows you to extract necessary results-based data from
#' different output files to create a MA plot (i.e. a scatter plot) of
#' \eqn{log_{2}} fold changes versus normalized mean counts while implementing
#' ggplot2 aesthetics.
#'
#' @param x treatment \code{x} for comparison (\eqn{log_{2}}(x/control)). This
#'  will be a factor level in your data.
#' @param y treatment \code{y} for comparison (\eqn{log_{2}}(y/control)). This
#'  will be a factor level in your data.
#' @param data output generated from calling the main routines of either
#'  \code{cuffdiff}, \code{DESeq2}, or \code{edgeR} analyses. For
#'  \code{cuffdiff}, this will be a \code{*_exp.diff} file. For \code{DESeq2},
#'  this will be a generated object of class \code{DESeqDataSet}. For
#'  \code{edgeR}, this will be a generated object of class \code{DGEList}.
#' @param d.factor a specified factor; for use with \code{DESeq2} objects only.
#'  This input equates to the first parameter for the contrast argument when
#'  invoking the \code{results()} function in \code{DESeq2}. Defaults to
#'  \code{NULL}.
#' @param type an analysis classifier to tell the function how to process the
#'  data. Must be either \code{cuffdiff}, \code{deseq}, or \code{edger}.
#'  \code{cuffdiff} must be used with \code{cuffdiff} data; \code{deseq} must
#'  be used for \code{DESeq2} output; \code{edgeR} must be used with
#'  \code{edgeR} data. See the \code{data} parameter for further details.
#' @param padj a user defined adjusted \code{p}-value cutoff point.
#'  Defaults to \code{0.05}.
#' @param y.lim set manual limits (boundaries) to the y axis. Defaults to
#'  \code{NULL}.
#' @param lfc \eqn{log} fold change level for setting conditonals. If no user
#'  input is added (\code{NULL}), value defaults to \code{1}.
#' @param title display the main title of plot. Logical; defaults to
#'  \code{TRUE}. If set to \code{FALSE}, no title will display in plot.
#' @param legend display legend of plot. Logical; defaults to \code{TRUE}.
#'  If set to \code{FALSE}, no legend will display in plot.
#' @param grid display major and minor axis lines. Logical; defaults to
#'  \code{TRUE}. If set to \code{FALSE}, no axis lines will display in plot.
#' @param highlight character string of IDs that will be highlighted. Set to
#'  \code{NULL} if you do not want highlighted data.
#' @param data.return returns data output of plot Logical; defaults to
#'  \code{FALSE}. If set to \code{TRUE}, a data frame will also be called.
#'  Assign to object for reproduction and saving of data frame. See final
#'  example for further details.
#' @param xaxis.title.size change the font size of the \code{x}-axis title
#'  text. Defaults to \code{10}.
#' @param xaxis.text.size change the font size of the \code{x}-axis text.
#'  Defaults to \code{10}.
#' @param yaxis.title.size change the font size of the \code{y}-axis title
#'  text. Defaults to \code{10}.
#' @param yaxis.text.size change the font size of the \code{y}-axis text.
#'  Defaults to \code{10}.
#' @param main.title.size change the font size of the plot title text.
#'  Defaults to \code{15}.
#' @param legend.text.size change the font size of the legend body text.
#'  Defaults to \code{9}.
#'
#' @return An object created by \code{ggplot}
#'
#' @export
#'
#' @examples
#' # Cuffdiff example
#' data("df.cuff")
#' vsMAPlot(
#'      x = "hESC", y = "iPS", data = df.cuff, d.factor = NULL,
#'      type = "cuffdiff", padj = 0.05, y.lim = NULL, lfc = 1,
#'      title = TRUE, legend = TRUE, grid = TRUE, data.return = FALSE
#' )
#'
#' # DESeq2 example
#' data("df.deseq")
#' require(DESeq2)
#' vsMAPlot(
#'      x = "treated_paired.end", y = "untreated_paired.end",
#'      data = df.deseq, d.factor = "condition", type = "deseq",
#'      padj = 0.05, y.lim = NULL, lfc = NULL, title = TRUE,
#'      legend = TRUE, grid = TRUE, data.return = FALSE
#' )
#'
#' # edgeR example
#' data("df.edger")
#' require(edgeR)
#' vsMAPlot(
#'      x = "WM", y = "MM", data = df.edger, d.factor = NULL,
#'      type = "edger", padj = 0.1, y.lim = NULL, lfc = 1,
#'      title = FALSE, legend = TRUE, grid = TRUE, data.return = FALSE
#' )
#'
#' # Highlight IDs
#' data("df.deseq")
#' require(DESeq2)
#' hl <- c(
#'     "FBgn0022201",
#'     "FBgn0003042",
#'     "FBgn0031957",
#'     "FBgn0033853",
#'     "FBgn0003371"
#' )
#' vsMAPlot(
#'      x = "treated_paired.end", y = "untreated_paired.end",
#'      data = df.deseq, d.factor = "condition", type = "deseq",
#'      padj = 0.05, y.lim = NULL, lfc = NULL, title = TRUE,
#'      legend = TRUE, grid = TRUE, data.return = FALSE, highlight = hl
#' )
#'
#' # Extract data frame from visualization
#' data("df.cuff")
#' tmp <- vsMAPlot(
#'      x = "hESC", y = "iPS", data = df.cuff,
#'      d.factor = NULL, type = "cuffdiff", padj = 0.05,
#'      y.lim = NULL, lfc = 1, title = TRUE, grid = TRUE,
#'      data.return = TRUE
#' )
#' df_ma <- tmp[[1]] ## or use tmp$data
#' head(df_ma)
#'
#' # Show plot from object (see prior example for more details)
#' tmp[[2]] ## or use tmp$plot

vsMAPlot <- function(
    x, y, data, d.factor = NULL, type = c("cuffdiff", "deseq", "edger"),
    padj = 0.05, y.lim = NULL, lfc = NULL, title = TRUE, legend = TRUE,
    grid = TRUE, highlight = NULL, data.return = FALSE, xaxis.text.size = 10,
    yaxis.text.size = 10, xaxis.title.size = 10, yaxis.title.size = 10,
    main.title.size = 15, legend.text.size = 9
) {
    if (missing(type) || !type %in% c("cuffdiff", "deseq", "edger")) {
        stop(
            paste(
                "Please specify analysis type",
                "(\"cuffdiff\", \"deseq\", or \"edger\")"
            )
        )
    }

    type <- match.arg(type)
    if (type == "cuffdiff") {
        dat <- .getCuffMA(x, y, data)
    } else if (type == "deseq") {
        dat <- .getDeseqMA(x, y, data, d.factor)
    } else if (type == "edger") {
        dat <- .getEdgeMA(x, y, data)
    }

    if (!isTRUE(title)) {
        m.lab <- NULL
    } else {
        m.lab  <- ggtitle(paste(y, "vs.", x))
    }

    if (!isTRUE(legend)) {
        leg <- theme(legend.position = "none")
    } else {
        leg <- guides(
            colour = guide_legend(
                override.aes = list(size = 3)
            ),
            shape = guide_legend(
                override.aes = list(size = 3)
            )
        )
    }

    if (!isTRUE(grid)) {
        grid <- theme_classic()
    } else {
        grid <- theme_bw()
    }

    dat$isDE <- ifelse(dat$padj <= padj, TRUE, FALSE)
    py <- dat$M

    if (is.null(y.lim)) {
        y.lim = c(-1.5, 1.5) * quantile(abs(py[is.finite(py)]), probs = 0.99)
    }
    if (is.null(lfc)) {
        lfc = 1
    }

    dat <- .ma.ranker(dat, padj, lfc, y.lim)

    tmp.size <- .ma.out.ranker(py, y.lim[2])
    tmp.col <- .ma.col.ranker(dat$isDE, py, lfc)
    tmp.shp <- .ma.shp.ranker(py, y.lim)

    tmp.cnt <- .ma.col.counter(dat, lfc)
    b <- tmp.cnt[[1]]
    g <- tmp.cnt[[2]]

    comp1 <- .ma.comp1(y.lim, padj, lfc, b, g)
    point <- geom_point(
        alpha = 0.7,
        aes(color = tmp.col, shape = tmp.shp, size = tmp.size)
    )
    comp2 <- .ma.comp2(
        comp1[[4]], comp1[[6]], comp1[[5]], comp1[[1]], comp1[[2]], comp1[[3]]
    )

    text.size <- theme(
        axis.text.x = element_text(size = xaxis.text.size),
        axis.text.y = element_text(size = yaxis.text.size),
        axis.title.x = element_text(size = xaxis.title.size),
        axis.title.y = element_text(size = yaxis.title.size),
        plot.title = element_text(size = main.title.size),
        legend.text = element_text(size = legend.text.size)
    )

    A <- id <- M <- NULL
    if (is.null(highlight)) {
        tmp.plot <- ggplot(
            dat, aes(x = A, y = pmax(y.lim[1], pmin(y.lim[2], py)))
        ) +
            point +
            comp2$color + comp2$shape + comp1$hline1 + comp1$hline2 +
            comp1$hline3 + comp1$x.lab + comp1$y.lab + m.lab + ylim(y.lim) +
            comp2$size + grid + leg + text.size
    } else {
        tl <- length(setdiff(highlight, dat$id))
        if (!is.atomic(highlight)) {
            stop("\"highlight\" must be vector.")
        } else if (all(highlight %in% dat$id)) {
            hl <- highlight
        } else if (tl > 0 && tl < length(highlight)) {
            remove <- setdiff(highlight, dat$id)
            message("Some IDs not found in data frame:")
            print(remove)
            message("Plotting the remaining samples...")
            hl <- highlight[!highlight %in% remove]
        } else if (!all(highlight %in% dat$id)) {
            stop("No IDs in highlight vector are present in data frame.")
        }

        A <- NULL
        tmp.plot <- ggplot(
            dat, aes(x = A, y = pmax(y.lim[1], pmin(y.lim[2], py)))
        ) +
            geom_point(
                alpha = 0.4,
                aes(color = tmp.col, shape = tmp.shp, size = tmp.size)
            ) +
            comp2$color + comp2$shape + comp1$hline1 + comp1$hline2 +
            comp1$hline3 + comp1$x.lab + comp1$y.lab + m.lab + ylim(y.lim) +
            comp2$size +
            ggrepel::geom_label_repel(
                data = dat[which(dat$id %in% hl), ],
                aes(label = id, x = A, y = M),
                segment.size = 1,
                segment.color = "gray10",
                box.padding = unit(0.4, "lines"),
                point.padding = unit(0.4, "lines")
            ) +
            geom_point(
                data = dat[which(dat$id %in% hl), ],
                aes(x = A, y = M),
                color = "red",
                size = 3
            ) +
            grid + leg + text.size
    }

    if (isTRUE(data.return)) {
        dat2 <- dat[, -ncol(dat)]
        plot.l <- list(data = dat2, plot = tmp.plot)
    } else {
        print(tmp.plot)
    }
}