ggPlotMzDelta

ggPlotMzDelta.R

ggPlotMzDelta <- function(delta, aaLabels = TRUE) {
    stopifnot(require("ggplot2"))
    ## from PSM::getAminoAcids()
    amino_acids <-
        structure(list(AA = c("peg", "A", "R", "N", "D", "C", "E", "Q",
                              "G", "H", "I", "L", "K", "M", "F", "P", "S",
                              "T", "W", "Y", "V"),
                       ResidueMass = c(44, 71.03711, 156.10111, 114.04293,
                                       115.02694, 103.00919, 129.04259,
                                       128.05858, 57.02146, 137.05891,
                                       113.08406, 113.08406, 128.09496,
                                       131.04049, 147.06841, 97.05276,
                                       87.03203, 101.04768, 186.07931,
                                       163.06333, 99.06841),
                       Abbrev3 = c(NA, "Ala", "Arg", "Asn", "Asp", "Cys",
                                   "Glu", "Gln", "Gly", "His", "Ile",
                                   "Leu", "Lys", "Met", "Phe", "Pro",
                                   "Ser", "Thr", "Trp", "Tyr", "Val"),
                       ImmoniumIonMass = c(NA, 44.05003, 129.114,
                                           87.05584, 88.03986, 76.0221,
                                           102.0555, 101.0715, 30.03438,
                                           110.0718, 86.09698, 86.09698,
                                           101.1079, 104.0534, 120.0813,
                                           70.06568, 60.04494, 74.06059,
                                           159.0922, 136.0762, 72.08133),
                       Name = c("Polyethylene glycol", "Alanine",
                                "Arginine", "Asparagine", "Aspartic acid",
                                "Cysteine", "Glutamic acid", "Glutamine",
                                "Glycine", "Histidine", "Isoleucine",
                                "Leucine", "Lysine", "Methionine",
                                "Phenylalanine", "Proline", "Serine",
                                "Threonine", "Tryptophan", "Tyrosine",
                                "Valine"),
                       Hydrophobicity = c(NA, 0.62, -2.53, -0.78, -0.9,
                                          0.29, -0.74, -0.85, 0.48, -0.4,
                                          1.38, 1.06, -1.5, 0.64, 1.19,
                                          0.12, -0.18, -0.05, 0.81, 0.26,
                                          1.08),
                       Hydrophilicity = c(NA, -0.5, 3, 0.2, 3, -1, 3, 0.2,
                                          0, -0.5, -1.8, -1.8, 3, -1.3,
                                          -2.5, 0, 0.3, -0.4, -3.4, -2.3,
                                          -1.5),
                       SideChainMass = c(NA, 15, 101, 58, 59, 47, 73, 72,
                                         1, 82, 57, 57, 73, 75, 91, 42,
                                         31, 45, 130, 107, 43),
                       pK1 = c(NA, 2.35, 2.18, 2.18, 1.88, 1.71, 2.19,
                               2.17, 2.34, 1.78, 2.32, 2.36, 2.2, 2.28,
                               2.58, 1.99, 2.21, 2.15, 2.38, 2.2, 2.29),
                       pK2 = c(NA, 9.87, 9.09, 9.09, 9.6, 10.78, 9.67,
                               9.13, 9.6, 8.97, 9.76, 9.6, 8.9, 9.21,
                               9.24, 10.6, 9.15, 9.12, 9.39, 9.11, 9.74),
                       pI = c(NA, 6.11, 10.76, 10.76, 2.98, 5.02, 3.08,
                              5.65, 6.06, 7.64, 6.04, 6.04, 9.47, 5.74,
                              5.91, 6.3, 5.68, 5.6, 5.88, 5.63, 6.02)),
                  class = "data.frame",
                  row.names = c(NA, -21L))
    ResidueMass <- ..density.. <- NULL ## to accomodate codetools
    value <- AA <- NULL
    delta <- data.frame(value = unlist(delta))
    p <- ggplot(delta, aes(x = value)) +
        geom_histogram(aes(y = ..density..), stat = "bin", binwidth = 1) +
        xlab("M/Z delta") + ylab("Density") +
        ggtitle("Histogram of Mass Delta Distribution")
    ## Adding annotations
    if (aaLabels) {
        y_offset <- x_offset <- rep(0.5, 21)
        names(y_offset) <- names(x_offset) <- amino_acids$AA
        x_offset[c("I", "L", "K", "Q")] <- 1
        y_offset[c("V", "C")] <- 1
        y_offset[c("P", "T")] <- 0
        y_offset[c("N", "E")] <- 1
        y_offset[c("K", "Q", "I", "L")] <- 0
        y_offset[c("D", "M")] <- 0
        aa <- cbind(amino_acids, x_offset, y_offset)
        ## removing Isoleucine, as it has the same residue mass
        ## as leucine, and updating leucine's label to I/L
        aa$AA <- as.character(aa$AA)
        aa[aa$AA == "I", "ResidueMass"] <- NA
        aa[aa$AA == "L", "AA"] <- "I/L"
        ## Removing Q as it is too close to K to show
        ## up correctly and updating K label to K/Q
        aa[aa$AA == "Q", "ResidueMass"] <- NA
        aa[aa$AA == "K", "AA"] <- "K/Q"
        p <- p +
            geom_vline(data = aa,
                       aes(xintercept = ResidueMass,
                           colour = AA),
                       alpha = I(1/2)) +
            geom_text(data = aa,
                      aes(x = ResidueMass,
                          y = -0.001, label = AA,
                          vjust = y_offset,
                          hjust = x_offset),
                      size = 2.5) +
            theme(legend.position = "none")
    }
    p
}