Upgrade to new Gitlab Version 13.9 on Saturday 19th April 20:00. Expect an interruption of about 30 to 60 minutes

auxfunc.R 11.4 KB
Newer Older
dmattek's avatar
dmattek committed
1
## Custom plotting
dmattek's avatar
dmattek committed
2 3
require(ggplot2)

dmattek's avatar
dmattek committed
4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
rhg_cols <- c(
  "#771C19",
  "#AA3929",
  "#E25033",
  "#F27314",
  "#F8A31B",
  "#E2C59F",
  "#B6C5CC",
  "#8E9CA3",
  "#556670",
  "#000000"
)

md_cols <- c(
  "#FFFFFF",
  "#F8A31B",
  "#F27314",
  "#E25033",
  "#AA3929",
  "#FFFFCC",
  "#C2E699",
  "#78C679",
  "#238443"
)

dmattek's avatar
dmattek committed
29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55
s.cl.linkage = c("ward.D",
                 "ward.D2",
                 "single",
                 "complete",
                 "average",
                 "mcquitty",
                 "centroid")

s.cl.spar.linkage = c("average",
                      "complete", 
                      "single",
                      "centroid")

s.cl.diss = c("euclidean", "maximum", "manhattan", "canberra", "binary", "minkowski")
s.cl.spar.diss = c("squared.distance","absolute.value")

l.col.pal = list(
  "White-Orange-Red" = 'OrRd',
  "Yellow-Orange-Red" = 'YlOrRd',
  "Reds" = "Reds",
  "Oranges" = "Oranges",
  "Greens" = "Greens",
  "Blues" = "Blues",
  "Spectral" = 'Spectral'
)


dmattek's avatar
dmattek committed
56 57 58 59 60 61 62 63 64 65 66 67
myGgplotTraj = function(dt.arg,
                        x.arg,
                        y.arg,
                        group.arg,
                        facet.arg,
                        facet.ncol.arg = 2,
                        line.col.arg = NULL,
                        xlab.arg = NULL,
                        ylab.arg = NULL,
                        plotlab.arg = NULL,
                        dt.stim.arg = NULL,
                        tfreq.arg = 1,
dmattek's avatar
dmattek committed
68
                        ylim.arg = NULL,
dmattek's avatar
dmattek committed
69 70 71 72
                        stim.bar.height.arg = 0.1,
                        stim.bar.width.arg = 0.5) {
  p.tmp = ggplot(dt.arg,
                 aes_string(x = x.arg,
dmattek's avatar
dmattek committed
73 74
                            y = y.arg,
                            group = group.arg))
dmattek's avatar
dmattek committed
75
  
dmattek's avatar
dmattek committed
76 77 78 79 80 81 82 83 84 85 86 87 88
  if (is.null(line.col.arg)) {
    p.tmp = p.tmp +
      geom_line(alpha = 0.25, 
                              size = 0.25)
  }
  else {
    p.tmp = p.tmp + 
      geom_line(aes_string(colour = line.col.arg), 
                              alpha = 0.5, 
                              size = 0.5) +
      scale_color_manual(name = '', 
                         values =c("FALSE" = rhg_cols[7], "TRUE" = rhg_cols[3], "SELECTED" = 'green', "NOT SEL" = rhg_cols[7]))
  }
dmattek's avatar
dmattek committed
89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114
  
  p.tmp = p.tmp + 
    stat_summary(
      aes_string(y = y.arg, group = 1),
      fun.y = mean,
      colour = 'blue',
      linetype = 'solid',
      size = 1,
      geom = "line",
      group = 1
    ) +
    facet_wrap(as.formula(paste("~", facet.arg)),
               ncol = facet.ncol.arg,
               scales = "free_x")
  
  if(!is.null(dt.stim.arg)) {
    p.tmp = p.tmp + geom_segment(data = dt.stim.arg,
                                 aes(x = Stimulation_time - tfreq.arg,
                                     xend = Stimulation_time - tfreq.arg,
                                     y = ylim.arg[1],
                                     yend = ylim.arg[1] + abs(ylim.arg[2] - ylim.arg[1]) * stim.bar.height.arg),
                                 colour = rhg_cols[[3]],
                                 size = stim.bar.width.arg,
                                 group = 1) 
  }
  
dmattek's avatar
dmattek committed
115 116 117
  if (!is.null(ylim.arg)) 
    p.tmp = p.tmp + coord_cartesian(ylim = ylim.arg)
  
dmattek's avatar
dmattek committed
118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146
  p.tmp = p.tmp + 
    xlab(paste0(xlab.arg, "\n")) +
    ylab(paste0("\n", ylab.arg)) +
    ggtitle(plotlab.arg) +
    theme_bw(base_size = 18, base_family = "Helvetica") +
    theme(
      panel.grid.minor = element_blank(),
      panel.grid.major = element_blank(),
      panel.border = element_blank(),
      axis.line.x = element_line(color = "black", size = 0.25),
      axis.line.y = element_line(color = "black", size = 0.25),
      axis.text.x = element_text(size = 12),
      axis.text.y = element_text(size = 12),
      strip.text.x = element_text(size = 14, face = "bold"),
      strip.text.y = element_text(size = 14, face = "bold"),
      strip.background = element_blank(),
      legend.key = element_blank(),
      legend.key.height = unit(1, "lines"),
      legend.key.width = unit(2, "lines"),
      legend.position = "top"
    )
  
  p.tmp
}


userDataGen <- function() {  
  cat(file=stderr(), 'userDataGen: in\n')
  
dmattek's avatar
dmattek committed
147
  locNtp = 40
dmattek's avatar
dmattek committed
148 149 150 151 152 153
  locNtracks = 5
  locNsites = 4
  locNwells = 2
  
  dt.nuc = data.table(Metadata_Site = rep(1:locNsites, each = locNtp * locNtracks),
                      Metadata_Well = rep(1:locNwells, each = locNtp * locNsites * locNtracks / locNwells),
dmattek's avatar
dmattek committed
154
                      Metadata_RealTime = rep(1:locNtp, locNsites* locNtracks),
dmattek's avatar
dmattek committed
155 156
                      objCyto_Intensity_MeanIntensity_imErkCor = c(rnorm(locNtp * locNtracks * locNsites * 0.5, .5, 0.1), rnorm(locNtp * locNtracks * locNsites * 0.5, 1, 0.2)),
                      objNuc_Intensity_MeanIntensity_imErkCor  = c(rnorm(locNtp * locNtracks * locNsites * 0.5, .25, 0.1), rnorm(locNtp * locNtracks * locNsites * 0.5, .5, 0.2)),
dmattek's avatar
dmattek committed
157 158 159 160 161
                      TrackLabel = rep(1:(locNtracks*locNsites), each = locNtp))
  
  cat(colnames(dt.nuc))
  return(dt.nuc)
}
dmattek's avatar
dmattek committed
162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220


# Returns original dt with an additional column with normalized quantity.
# The column to be normalised is given by 'in.meas.col'.
# The name of additional column is the same as in.meas.col but with ".norm" suffix added.
# Normalisation is based on part of the trajectory;
# this is defined by in.rt.min and max, and the column with time in.rt.col.
# Additional parameters:
# in.by.cols - character vector with 'by' columns to calculate normalisation per group
#              if NULL, no grouping is done
# in.robust - whether robust measures should be used (median instead of mean, mad instead of sd)
# in.type - type of normalization: z.score or mean (fi.e. old change w.r.t. mean)

myNorm = function(in.dt,
                  in.meas.col,
                  in.rt.col = 'RealTime',
                  in.rt.min = 10,
                  in.rt.max = 20,
                  in.by.cols = NULL,
                  in.robust = TRUE,
                  in.type = 'z.score') {
  loc.dt <-
    copy(in.dt) # copy so as not to alter original dt object w intermediate assignments
  
  if (is.null(in.by.cols)) {
    if (in.robust)
      loc.dt.pre.aggr = loc.dt[get(in.rt.col) > in.rt.min &
                                 get(in.rt.col) < in.rt.max, .(meas.md = median(get(in.meas.col), na.rm = TRUE),
                                                               meas.mad = mad(get(in.meas.col), na.rm = TRUE))]
    else
      loc.dt.pre.aggr = loc.dt[get(in.rt.col) > in.rt.min &
                                 get(in.rt.col) < in.rt.max, .(meas.md = mean(get(in.meas.col), na.rm = TRUE),
                                                               meas.mad = sd(get(in.meas.col), na.rm = TRUE))]
    
    loc.dt = cbind(loc.dt, loc.dt.pre.aggr)
  }  else {
    if (in.robust)
      loc.dt.pre.aggr = loc.dt[get(in.rt.col) > in.rt.min &
                                 get(in.rt.col) < in.rt.max, .(meas.md = median(get(in.meas.col), na.rm = TRUE),
                                                               meas.mad = mad(get(in.meas.col), na.rm = TRUE)), by = in.by.cols]
    else
      loc.dt.pre.aggr = loc.dt[get(in.rt.col) > in.rt.min &
                                 get(in.rt.col) < in.rt.max, .(meas.md = mean(get(in.meas.col), na.rm = TRUE),
                                                               meas.mad = sd(get(in.meas.col), na.rm = TRUE)), by = in.by.cols]
    
    loc.dt = merge(loc.dt, loc.dt.pre.aggr, by = in.by.cols)
  }
  
  
  if (in.type == 'z.score') {
    loc.dt[, meas.norm := (get(in.meas.col) - meas.md) / meas.mad]
  } else {
    loc.dt[, meas.norm := (get(in.meas.col) / meas.md)]
  }
  
  setnames(loc.dt, 'meas.norm', paste0(in.meas.col, '.norm'))
  
  loc.dt[, c('meas.md', 'meas.mad') := NULL]
  return(loc.dt)
dmattek's avatar
dmattek committed
221 222
}

dmattek's avatar
dmattek committed
223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321
# Plots a scatter plot with marginal histograms
# Points are connected by a line (grouping by cellID)
#
# Assumes an input of data.table with
# x, y - columns with x and y coordinates
# id - a unique point identifier (here corresponds to cellID)
# mid - a (0,1) column by which points are coloured (here corresponds to whether cells are within bounds)

myGgplotScat = function(dt.arg,
                        band.arg = NULL,
                        facet.arg = NULL,
                        facet.ncol.arg = 2,
                        xlab.arg = NULL,
                        ylab.arg = NULL,
                        plotlab.arg = NULL,
                        alpha.arg = 1,
                        group.col.arg = NULL) {
  p.tmp = ggplot(dt.arg, aes(x = x, y = y))
  
  if (is.null(group.col.arg)) {
    p.tmp = p.tmp +
      geom_point(alpha = alpha.arg, aes(group = id))
  } else {
    p.tmp = p.tmp +
      geom_point(aes(colour = as.factor(get(group.col.arg)), group = id), alpha = alpha.arg) +
      geom_path(aes(colour = as.factor(get(group.col.arg)), group = id), alpha = alpha.arg) +
      scale_color_manual(name = group.col.arg, values =c("FALSE" = rhg_cols[7], "TRUE" = rhg_cols[3], "SELECTED" = 'green'))
  }
  
  if (is.null(band.arg))
    p.tmp = p.tmp +
      stat_smooth(
        method = function(formula, data, weights = weight)
          rlm(formula, data, weights = weight, method = 'MM'),
        fullrange = FALSE,
        level = 0.95,
        colour = 'blue'
      )
  else {
    p.tmp = p.tmp +
      geom_abline(slope = band.arg$a, intercept = band.arg$b) +
      geom_abline(
        slope = band.arg$a,
        intercept =  band.arg$b + abs(band.arg$b)*band.arg$width,
        linetype = 'dashed'
      ) +
      geom_abline(
        slope = band.arg$a,
        intercept = band.arg$b - abs(band.arg$b)*band.arg$width,
        linetype = 'dashed'
      )
  }
  
  if (!is.null(facet.arg)) {
    p.tmp = p.tmp +
      facet_wrap(as.formula(paste("~", facet.arg)),
                 ncol = facet.ncol.arg)
    
  }
  
  
  if (!is.null(xlab.arg))
    p.tmp = p.tmp +
      xlab(paste0(xlab.arg, "\n"))
  
  if (!is.null(ylab.arg))
    p.tmp = p.tmp +
      ylab(paste0("\n", ylab.arg))
  
  if (!is.null(plotlab.arg))
    p.tmp = p.tmp +
      ggtitle(paste0(plotlab.arg, "\n"))
  
  
  
  p.tmp = p.tmp +
    theme_bw(base_size = 18, base_family = "Helvetica") +
    theme(
      panel.grid.minor = element_blank(),
      panel.grid.major = element_blank(),
      axis.line.x = element_line(color = "black", size = 0.25),
      axis.line.y = element_line(color = "black", size = 0.25),
      axis.text.x = element_text(size = 12),
      axis.text.y = element_text(size = 12),
      strip.text.x = element_text(size = 14, face = "bold"),
      strip.text.y = element_text(size = 14, face = "bold"),
      strip.background = element_blank(),
      legend.key = element_blank(),
      legend.key.height = unit(1, "lines"),
      legend.key.width = unit(2, "lines"),
      legend.position = "none"
    )
  
  # Marginal distributions don;t work with plotly...
  # if (is.null(facet.arg))
  #   ggExtra::ggMarginal(p.scat, type = "histogram",  bins = 100)
  # else
  return(p.tmp)
}
dmattek's avatar
dmattek committed
322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338

myGgplotTheme = theme_bw(base_size = 18, base_family = "Helvetica") +
  theme(
    panel.grid.minor = element_blank(),
    panel.grid.major = element_blank(),
    axis.line.x = element_line(color = "black", size = 0.25),
    axis.line.y = element_line(color = "black", size = 0.25),
    axis.text.x = element_text(size = 12, angle = 45, hjust = 1),
    axis.text.y = element_text(size = 12),
    strip.text.x = element_text(size = 14, face = "bold"),
    strip.text.y = element_text(size = 14, face = "bold"),
    strip.background = element_blank(),
    legend.key = element_blank(),
    legend.key.height = unit(1, "lines"),
    legend.key.width = unit(2, "lines"),
    legend.position = "right"
  )