R: Using the segments function to plot a map of stacked lines

问题

I have written a function that plots a number of lines along an axis, stacking them where they overlap. Below is the code, a sample table, and the image that it produces.

The plot is mostly exactly what I was looking for but for a few things (in order of importance):

Plotting the segments is an extremely slow process: about 1 segment every 0.5 seconds. Considering they are just lines I expected something much faster. I don't know the cause of this. I know that explicit loops can be slow in R so it might be this, or should I be plotting off-screen somehow and then pushing the plot to screen afterwards? Finding a time efficient method for plotting this kind of map is important because my tables can easily be tens of thousands of rows long.
I can't find any way of specifying the gap between y positions to be a fixed distance regardless of the number of Y positions. At the extreme, plotting just two segments produces a plot with the segments very far apart from each other.

Can anyone help me with either of these points (or indeed, anything else I could be doing better)?

(In this code reads == segments)

The function:

viewReads <- function(reads){
    # sort by start
    sorted <- reads[order(reads$start),];

    #---
    # In the first iteration we work out the y-axis
    # positions that segments should be plotted on
    # segments should be plotted on the next availible
    # y position without merging with another segment
    #---
    yread <- c(); #keeps track of the x space that is used up by segments 

    # get x axis limits
    minstart <- min(sorted$start);
    maxend <- max(sorted$end);

    # initialise yread
    yread[1] <- minstart - 1;
    ypos <- c(); #holds the y pos of the ith segment

    # for each read
    for (r in 1:nrow(sorted)){
        read <- sorted[r,];
        start <- read$start;
        placed <- FALSE;

        # iterate through yread to find the next availible
        # y pos at this x pos (start)
        y <- 1;
        while(!placed){

            if(yread[y] < start){
                ypos[r] <- y;
                yread[y] <- read$end;
                placed <- TRUE;
            } 

            # current y pos is used by another segment, increment
            y <- y + 1;
            # initialize another y pos if we're at the end of the list
            if(y > length(yread)){
                yread[y] <- minstart-1;
            }
        }
    } 

    # find the maximum y pos that is used to size up the plot
    maxy <- length(yread);
    sorted$ypos <- ypos;

    # Now we have all the information, start the plot
    plot.new();
    plot.window(xlim=c(minstart, maxend+((maxend-minstart)/10)), ylim=c(1,maxy));
    axis(3);

    #---
    # This second iteration plots the segments using the found y pos and 
    # the start and end values
    #---
    for (r in 1:nrow(sorted)){
        read <- sorted[r,];
        # colour dependent on strand type
        if(read$strand == '+'){
            color = 'blue'
         }else{
            color = 'red'
         } 
        #plot this segment!
        segments(read$start, maxy-read$ypos, read$end, maxy-read$ypos, col=color);
    }
}

Sample code:

start   end strand
86  115 +
87  115 +
91  116 +
88  117 +
91  117 +
98  125 -
104 131 +
104 131 +
106 132 -
104 134 +
104 134 +
104 134 +
106 134 +
106 134 +
106 134 +
106 134 +
106 134 +
106 135 +
106 135 +
106 135 +
106 135 +
106 135 +
106 135 +
106 135 +
108 135 +
108 135 +
108 135 +
108 135 +
108 135 +
108 135 +
108 135 +
108 135 +
108 135 +
108 135 +
108 135 +
108 135 +
108 135 +
108 135 +
108 135 +
108 135 +
108 135 +
108 135 +
108 135 +
108 135 +
109 135 +
116 135 -
106 136 +
106 136 +
106 136 +
108 136 +
108 136 +
108 136 +
108 136 +
108 136 +
108 136 +
108 136 +
108 136 +
108 136 +
108 137 +
108 137 +
109 137 -
108 138 +
108 138 +
108 138 +
108 138 +
112 138 +
112 139 +
119 141 +
116 143 +
121 145 +
127 145 -
119 146 +
121 148 +
142 169 -
142 169 -
160 185 -
162 185 -
165 185 -

result:

回答1:

Sorry, I don't have time for a worked-through example, but segments() (as well as other functions, like polygons(), points(), etc can take their arguments as vectors, such that you can do all of your drawing in a single function call. Often preparing the arguments prior to plotting (apply()ing or looping if necessary) can be way faster than repeated calls to these drawing functions. The answer in this post: Plotting a rather complex chart using R and axis break() gives a complete example of this approach. You'll definitely be able to apply this to your situation. Good luck! (and thanks for telling me to answer)

来源：https://stackoverflow.com/questions/9607527/r-using-the-segments-function-to-plot-a-map-of-stacked-lines

标签

plot

segments