Cropping the top and bottom output of ggsave from grid.arrange of PNG Files - r

library(grid)
library(gridExtra)
library(png)
library(ggplot2)
library(RCurl)
PNG_1 <- readPNG(getURLContent("https://i.ibb.co/MVx1QsQ/A.png"))
PNG_2 <- readPNG(getURLContent("https://i.ibb.co/kHVGNfQ/B.png"))
PNG_3 <- readPNG(getURLContent("https://i.ibb.co/yVf3Hjg/C.png"))
grid <- grid.arrange(rasterGrob(PNG_1), rasterGrob(PNG_2), rasterGrob(PNG_3), ncol=3)
ggsave(grid,filename="output.png")
Output.png
Tried to set the output dimensions manually, but to no avail.
Simply wish to remove the large top and bottom margins. Thanks.


As plots adapts to device size it's not simple to get part size. I think the easiest way is to compute aspect ratio from dimension of PNGs and provide device size to ggsave.
asp <- (ncol(PNG_1)+ncol(PNG_2)+ncol(PNG_3))/max(nrow(PNG_1),nrow(PNG_2),nrow(PNG_3))
ggsave(grid, filename= ..., width=5, height=5/asp)
to stitch those PNG you can use this code that will keep original pixel count.
We can't use cbind/rbind function as readPNG returns arrays and not matrices. Fortunately the abind package provide a function to do it, abind(
along=2 means cbind, along=1 for rbind)
library(abind)
PNG <- abind(PNG_1,PNG_2,PNG_3,along=2)
writePNG(PNG,"output.png")

Related

R - How to change size of a whole figure produced by grid.arrange?

I want to display multiple ggplot objects in one figure using package gridExtra. It works, but titles are cut. I can fix it by changing plots font size, but I would rather change the width of the whole figure so that it is wide enough to fit in titles (and as wide as other outputs in my RMarkdown doc, e.g. the table right over it). I've tried to change margins by typing par(mai = 2*par('mai')) before call to grid.arrange, but it did nothing. Does anyone know how to do it properly? I am really confused about the gridExtra package.
library(gridExtra)
g <- grid.arrange(zad4_kl$pw, zad4_kl$pk, zad4_kl$pp, zad4_AC$pw, zad4_AC$pk, zad4_AC$pp, nrow = 2)
g
Screenshot of produced output

If you are saving the image, then you can increase the width and dpi in ggsave.
In R Markdown, you can use fig.width and fig.height
{r, echo=FALSE, fig.width=8, fig.cap="A nice image."}
library(gridExtra)
g <- grid.arrange(zad4_kl$pw, zad4_kl$pk, zad4_kl$pp, zad4_AC$pw, zad4_AC$pk, zad4_AC$pp, nrow = 2)
g
If rendered image doesn't fit the Markdown output, then you will have to change font in individual plots.

How to remove background of images with the magick package?

Is it possible to remove the background of images with the magick package?
I know how to use edge detection with Gimp/Inkscape to crop out silhouettes; however, I'm looking to automate the process for a large batch of images with R.
My ultimate goal is to use the ggimage package to plot these images as x,y coordinates but the background of these images is currently overlapping the plot (the dog compared to fink)
library("ggplot2")
library("ggimage")
set.seed(2017-02-21)
d <- data.frame(x = rnorm(10),
y = rnorm(10),
image = sample(c("http://www.supercoloring.com/sites/default/files/silhouettes/2015/05/cairn-terrier-black-silhouette.svg", "https://jeroenooms.github.io/images/frink.png"),
size=10, replace = TRUE)
)
ggplot(d, aes(x, y)) + geom_image(aes(image=image))
One can trim the edges of an image using image magick's image_trim()
img <- image_read_svg("http://www.supercoloring.com/sites/default/files/silhouettes/2015/05/cairn-terrier-black-silhouette.svg")
image_trim(img)
but this isn't exactly what I would like.
Any ideas?

The image_transparent() function does this, e.g.:
logo <- image_read("logo:")
image_transparent(logo, 'white')
The white parts of the image will be made transparent, which should be enough for the simple image in your example.

see my answer in https://yulab-smu.github.io/treedata-book/image-processing-using-magick-package.html#example-1-remove-background-of-images.
You can pass image processing function provided by magick package to geom_image via the image_fun parameter.

R figure being cropped

Here's a bit of a basic question, you'd think R would automate this always. I'm making a heatmap (basic stuff, png() into heatmap.2() into dev.off()), and my row labels get cleaved off on the right hand side, along with a tiny part of the dendrogram on the left:
Fiddling with the margins manually would eventually lead to a sensible output, but the problem is that this is part of an automated pipeline with biologist users in mind. Surely there's some automated way to do this, right? I mean, in RGui on OSX, a quick reshape of the figure window fixes the layout automatically, so it has to be possible. Thanks, and sorry to bother you.
EDIT: Procedure to reproduce this visualisation:
Input data:
Time,2,4,6,8,10,12,14,16,18,20,22,24,26,28,30,32,34,36,38,40,42,44,46,48,2,4,6,8,10,12,14,16,18,20,22,24,26,28,30,32,34,36,38,40,42,44,46,48,2,4,6,8,10,12,14,16,18,20,22,24,26,28,30,32,34,36,38,40,42,44,46,48,2,4,6,8,10,12,14,16,18,20,22,24,26,28,30,32,34,36,38,40,42,44,46,48
CATMA1A23000,0.21129,0.48322,0.54234,0.93714,0.51816,0.11103,0.38304,0.97183,0.61381,-0.13775,0.32575,0.85221,0.1921,1.8692,0.4117,0.51977,0.74019,0.98932,0.023536,0.8849,0.75141,0.42846,0.03839,0.50295,0.38849,-0.95087,0.55017,-0.14813,0.79036,0.40978,-0.33048,0.81173,0.60585,0.014413,0.86586,0.063947,1.1195,0.25425,-0.090325,1.3035,0.90609,0.62866,0.83168,0.88655,0.4843,0.60098,0.26669,0.66605,0.78468,-0.76709,0.29615,-0.8911,0.5022,-0.93834,-0.065904,0.58003,0.41217,-0.99012,-0.032054,0.69119,1.1791,-0.0059458,0.81002,0.76421,1.302,0.77637,0.7824,1.192,-0.27005,-0.24698,1.8706,0.33513,0.38592,1.1987,0.48569,0.7104,2.4307,0.68148,0.11407,0.65619,0.94468,1.245,0.44146,0.71048,0.96389,1.7647,0.22729,0.96875,-0.2738,1.2975,2.0503,0.52507,2.282,1.0773,0.44489,0.22504
CATMA1A30870,3.9909,5.1844,5.2522,5.5482,5.665,5.4666,5.4456,5.5456,5.0014,5.5408,3.9324,3.9288,4.3624,3.0828,5.4821,5.1446,4.5638,4.6449,5.2604,4.2607,4.0942,3.9892,5.1361,3.8625,3.9454,5.3148,5.188,6.088,5.2085,5.3073,5.5652,4.8358,4.6111,4.9911,3.1497,4.3796,4.2456,3.9712,5.0998,5.4383,5.0908,4.5155,4.6486,5.1233,4.4259,4.7218,3.7928,4.0384,4.1085,6.2787,5.4937,5.9637,5.4085,5.188,5.3057,5.1963,4.3807,4.8352,3.9584,3.5859,3.8535,3.9755,4.5199,5.1093,5.3526,4.1214,4.5787,3.4499,4.1962,4.6929,4.3379,4.6376,4.4774,5.2341,5.3061,5.6769,3.6244,5.1073,4.543,4.9929,4.9341,3.9829,4.4304,4.0021,3.7498,4.2972,5.3338,4.8094,4.5766,5.5171,3.1207,5.0926,3.0253,4.351,4.351,4.5253
CATMA1A65800,0.65615,0.54186,0.93488,0.2205,0.38134,0.95061,0.60157,0.66508,0.83971,1.3625,0.9037,1.1606,1.0143,0.91896,1.3587,1.2468,1.0147,0.44113,1.5196,0.60232,0.52456,0.82557,0.43697,1.3253,0.78614,0.73697,1.0593,0.010271,0.59042,0.58544,0.38307,1.0372,1.071,0.21915,1.1779,0.97341,0.95672,0.92057,1.3866,1.3884,0.99241,1.0546,1.0215,1.0144,1.1606,1.2421,0.89549,0.56747,0.99347,1.0706,1.3403,0.48719,0.30759,1.0215,0.76474,0.87313,0.47597,1.5028,1.2201,1.7923,0.64299,1.3612,1.0079,0.64145,1.0969,0.90254,0.9109,0.85213,0.55241,0.63797,1.5373,0.45586,0.59624,0.5717,0.13919,-0.012128,2.7367,0.83577,0.13422,0.13908,0.32377,0.013345,-0.017209,0.47044,1.034,1.4932,1.0067,0.57515,1.1405,0.95364,3.0214,0.14601,2.1645,1.1704,0.45205,0.83406
CATMA1C71184,4.2879,2.7256,3.306,2.2375,2.6575,4.6438,4.3899,3.8518,5.3494,3.8049,5.4095,4.5437,3.4362,4.0983,2.5059,2.7275,3.9432,3.5177,4.266,4.6446,4.9338,3.5188,3.7903,4.4867,3.9692,3.3382,3.2898,3.4103,3.3025,4.1279,4.3532,5.0625,5.756,4.7799,5.0303,4.8974,3.8698,2.887,2.7541,2.7837,3.0665,4.0168,3.7507,4.6273,5.2384,3.7049,4.5843,3.97,2.7142,3.6205,3.5394,3.3561,3.2672,4.4467,5.2044,5.2177,3.8496,5.7881,4.7594,5.0395,2.8281,2.8751,2.8322,2.6873,1.8326,3.042,3.834,4.7844,3.1947,3.2571,4.0826,4.8398,3.8147,2.7439,2.846,2.64,2.8055,3.421,3.8495,4.9367,4.6413,3.5281,4.0636,3.7144,3.4468,2.3466,1.7452,2.3151,4.022,2.9211,3.89,3.7213,4.1856,3.3613,4.0891,4.2135
CATMA1C71776,-2.0266,-0.77692,-1.3279,-0.40422,-0.38614,-0.34387,-0.44277,0.99775,1.0748,0.17831,-0.14209,-0.98905,-0.88619,0.40548,-0.37731,0.18112,-0.80473,-0.26957,0.49049,0.25413,0.30341,0.20203,-1.0343,-0.17877,-0.46107,-1.202,-0.58841,-1.0995,-1.1079,-0.10977,0.29428,0.33363,0.35537,0.55571,-0.11074,-0.69584,-0.052035,-0.18207,-0.45321,-0.2778,-0.81074,-0.27129,0.087758,0.52214,0.33791,0.51601,-0.30863,0.14386,-0.2922,-0.21013,-0.73268,-1.1732,-0.36599,0.4731,1.0357,0.62377,0.8219,0.6029,-0.27195,1.0677,-0.48423,-0.57601,-0.43393,-2.0519,-0.67433,-0.44234,0.0078619,0.40677,-0.47102,-0.59971,0.30058,-0.4604,-0.56523,-3.3913,-0.52856,-0.54137,1.0418,-0.26115,0.21246,0.37858,0.38716,-0.86708,-0.41297,-0.21536,-0.06004,-0.48526,-1.2651,-0.52592,0.23757,-0.73376,1.0272,0.59383,0.65704,-0.017658,-0.44128,-0.30226
CATMA2A41800,1.5057,1.8357,2.7851,2.9125,2.6421,2.3881,2.4772,2.3367,2.2988,3.0073,2.7659,2.8454,2.5939,2.6327,2.6184,2.3415,3.3473,2.7683,1.1123,1.9614,1.8795,2.2335,2.9227,3.1536,2.6255,2.5264,2.3874,3.3196,1.8572,2.3446,2.3358,1.683,2.1742,2.8082,2.8634,2.8607,2.6755,3.2567,2.0726,3.0746,2.8984,2.5235,2.9399,2.5731,1.9196,2.9845,2.965,2.8055,1.1878,-0.6181,2.9552,2.7814,3.0701,-0.42372,2.5897,2.6331,1.7218,-0.70788,2.9578,1.1506,3.3772,1.6571,3.3166,3.5001,2.6541,2.5367,2.8643,2.1034,2.6389,1.8915,2.4861,3.2271,1.7718,2.7359,2.391,2.7223,1.1594,2.4025,1.1524,2.1405,2.253,2.4368,2.8053,2.2955,2.2877,2.7792,3.4115,3.2464,-0.37629,3.0582,1.682,2.0598,1.8747,2.0158,2.4714,2.1816
CATMA2C47274,2.2048,1.6574,0.96502,0.15092,0.5251,0.40351,0.90106,-0.12415,0.48846,0.03794,0.86292,0.27027,0.71848,1.8535,0.47188,-0.067504,0.68578,0.28113,0.17559,0.29659,0.067618,0.79368,-0.36732,0.80945,2.4297,1.3272,1.3741,0.005024,0.57369,0.6195,0.17845,1.3991,0.71594,-0.76358,0.74421,0.078431,0.97596,0.14151,-0.77085,0.64926,-0.24602,1.3478,0.49879,0.90549,0.78024,0.94685,0.023589,0.07446,1.3597,0.19311,1.1883,-0.20756,0.59042,0.031659,0.3591,1.1757,0.97314,-1.0133,0.92409,1.2264,-0.0095851,-0.47315,1.3107,1.3967,0.80332,0.5494,1.0394,1.4519,0.50253,-0.65282,1.5689,0.17996,1.5938,0.28124,0.93511,0.10344,3.6906,0.56448,-0.40754,0.75218,0.10346,-0.37737,-0.53789,0.57109,0.29602,-0.64821,-0.52805,0.34899,0.50146,0.61456,2.1785,-0.41,2.135,0.7426,0.3307,0.64542
CATMA2C47710,3.9021,3.7272,3.709,3.494,3.7557,3.7883,3.8697,3.4952,3.6365,3.6948,4.0954,3.7629,4.1282,3.6108,4.2769,3.3689,3.5232,3.7488,3.6237,4.0805,3.855,3.8724,4.2706,3.5584,3.9576,4.1324,3.857,3.6508,3.7609,3.7885,3.6256,3.8022,3.7189,3.7417,3.6357,3.6448,4.2841,3.8731,3.6047,3.7792,3.5703,3.8284,3.8098,3.8824,4.2441,3.6127,3.8253,3.529,3.6078,4.0917,3.6794,3.7977,3.8231,3.9369,3.6087,3.7634,3.6419,4.0052,3.7083,4.0191,3.3211,3.9133,3.74,4.0801,4.0999,3.6312,3.7747,3.7768,3.7977,3.9409,3.9423,3.6708,4.1773,3.666,3.8267,3.8832,4.0462,3.8007,3.3932,4.0945,3.9901,4.1036,4.1064,3.8178,3.7936,3.8638,4.3395,3.5859,3.5698,3.7885,4.4263,4.0292,3.6857,4.1197,4.0262,3.7753
CATMA3A13690,5.2022,5.6481,5.7765,4.8253,6.0581,6.2026,5.2711,6.4481,6.0929,6.3395,5.3806,4.9156,5.3634,4.5556,5.3547,5.5214,5.6947,5.9289,5.6269,5.5406,6.0508,5.8498,5.6275,5.4422,5.3293,5.7543,5.1577,6.3626,5.429,5.7245,5.7554,5.4575,5.2244,6.2312,6.0425,5.0317,5.4494,5.5252,5.5124,5.6588,5.0015,5.7454,5.6711,6.132,5.4789,6.1668,5.6978,5.113,4.3572,5.6177,5.9557,6.0099,5.2522,5.5598,6.2427,5.4703,5.4275,5.9159,5.3029,4.1094,5.927,5.9223,5.1257,5.3835,5.6579,5.5849,5.8674,4.9425,5.2144,5.5382,4.7165,5.5688,5.2766,6.4298,5.9438,5.7105,4.757,5.9097,5.3665,5.8336,5.8393,5.6104,5.4702,5.4028,5.0756,5.632,5.9963,5.7889,5.7521,5.5561,4.0685,6.3026,4.9628,6.176,5.7256,4.8529
CATMA3A15560,0.23073,0.32331,0.37953,0.081038,-0.33446,0.18495,0.86901,0.71363,0.33531,1.4741,0.32824,0.18869,-0.34772,0.97628,0.62455,-0.14645,0.42603,-0.00031328,0.92441,0.14199,0.25522,0.55583,0.10067,1.2818,0.64424,-0.24454,0.33039,0.25975,1.1023,-0.5464,0.62174,-0.15397,0.78478,0.2999,-0.063974,0.55156,0.44076,0.17798,0.61707,0.9308,0.19809,1.2378,0.26328,0.19767,0.069484,0.76748,0.45386,0.4495,0.050625,0.99311,-0.10988,-0.070507,0.46274,0.19475,0.36696,-0.16537,1.0566,0.94446,0.85272,1.194,-0.41335,0.17202,0.38729,0.65495,-0.32063,0.60186,0.24444,1.3128,0.48551,0.55998,1.8149,0.36494,0.10693,-0.19085,0.25082,-0.10628,-0.38432,0.50117,0.97923,-0.31536,-0.2634,-0.82767,0.70017,-0.0093473,0.27653,0.67773,0.40978,0.3315,0.5179,0.32044,2.9356,-0.51529,2.7438,0.4153,0.32022,0.74311
CATMA3A21390,-0.54603,-0.32262,-0.29966,-0.22463,-0.49654,-1.0552,-1.0705,-0.93928,-0.71247,-2.7228,-0.67849,-0.24868,-1.5015,-0.22486,-0.26215,-0.34298,-0.73168,-0.33699,-0.3455,-0.5349,-0.68784,-0.47506,-1.2541,0.86942,-0.14503,-2.0603,-0.35647,-0.67731,0.20243,-1.1881,-1.3828,-1.0471,-0.20157,-2.5492,-0.7862,-0.21715,-0.20583,-0.067758,-0.86589,-0.75509,-0.62777,-0.57176,-2.0996,-0.72894,-0.56391,-1.1054,-0.94172,-0.98805,-0.20763,-0.42764,-2.0142,-1.4079,-0.66326,-0.10354,-1.6284,-1.9413,-0.77001,-0.58492,-0.99846,-0.013285,-0.31205,-0.10665,-1.2396,-1.7556,-0.40454,-0.26631,-1.2694,-0.32263,-1.2293,-0.93683,0.97994,-1.8696,-0.26859,-0.25661,-0.33193,-0.83889,-0.25944,-0.63826,0.3387,-1.4443,-0.93123,-0.84915,-0.24606,-0.63109,-0.78621,-0.40802,-0.6463,-0.84319,-1.0379,-1.1471,1.9932,-1.3029,1.5286,-0.19754,0.018446,-0.55479
CATMA3A53880,3.917,3.6689,3.7213,3.8812,3.8253,4.4861,4.258,4.7926,4.9737,4.0869,3.2183,3.5812,3.408,3.3128,3.3915,4.1923,3.9122,4.0586,4.2707,4.3077,4.8926,4.1982,3.4817,3.5267,3.6952,3.6632,3.9304,4.0793,3.9499,4.2175,4.8303,4.9599,4.6336,3.8998,3.7932,3.8572,3.2167,3.7749,4.3971,4.0662,3.9241,4.6407,4.2837,4.9644,4.972,4.2565,3.8796,3.383,3.2382,4.1017,4.058,4.0423,3.6033,4.0774,5.141,4.9736,4.4372,3.8969,3.7386,3.3618,3.539,4.2588,3.4726,3.7424,3.5818,4.9159,4.3567,4.1291,4.1108,4.4035,3.1374,3.4675,3.762,3.9264,3.7319,4.1205,2.0915,4.4192,4.5049,4.5427,4.6205,3.4653,3.6017,3.3376,3.266,3.5561,3.4538,3.761,4.2013,4.1787,4.017,4.7773,3.748,4.4977,3.8685,3.7464
CATMA5A09500,6.2838,5.1076,6.6209,5.7153,5.9395,5.3601,4.2366,5.189,3.3074,4.8701,5.0357,5.0951,6.2843,5.691,5.5054,5.9949,5.0834,5.1484,4.5884,4.805,4.458,5.0816,4.7264,5.3661,5.0645,5.8293,5.2877,5.6602,5.3336,4.7329,4.843,5.1901,3.8431,4.683,5.7106,4.0774,6.5689,5.8061,5.3445,5.2905,3.837,4.6271,4.2277,4.6165,4.0081,4.5176,4.259,4.6363,4.8557,3.9828,5.9728,5.4698,5.4102,4.5802,5.0247,4.8967,4.0116,2.6636,5.4684,4.3454,6.6893,6.5624,6.1464,5.4444,5.3309,5.6682,5.5012,4.3162,2.9028,3.3906,4.6684,4.914,5.3601,5.5138,6.2191,5.8043,5.8165,5.1155,3.8681,4.9268,4.0232,6.1699,4.6327,5.7364,6.1934,6.7188,5.0755,5.7058,5.3703,4.3921,3.2108,4.9503,4.1003,4.8225,4.6701,4.7166
CATMA5A12680,1.7891,1.362,1.4303,0.61973,0.54124,1.2733,0.7775,0.66788,1.7338,0.63125,1.5852,0.91616,1.0712,0.81919,0.90152,0.77799,1.0835,0.32735,2.0868,0.5895,1.1445,0.64545,0.71776,0.33929,1.42,1.0729,1.4879,0.86379,0.95453,1.2928,0.97753,1.1738,0.79397,0.31563,0.86345,0.67595,1.3501,1.079,1.2034,1.2432,1.3696,1.1473,0.72528,0.57199,0.85965,0.65339,0.47614,0.43219,1.363,2.4026,0.98947,0.85204,1.2591,2.3151,1.6775,1.6555,1.0371,2.5586,-0.097071,0.91744,1.068,1.6819,1.0122,0.99911,0.75267,1.2195,0.42175,0.2554,0.4819,0.80025,0.44517,0.32442,1.9863,1.5019,1.2893,1.0669,1.6881,1.7581,1.6505,1.943,2.141,1.018,0.68207,0.84392,1.1452,1.009,1.1376,1.0572,1.9258,1.7039,0.10261,1.4549,0.94693,0.77945,0.049658,0.84178
CATMA5A14990,0.41247,0.51378,0.375,-0.24526,0.027658,-0.015345,-0.028278,-0.2643,0.3905,0.12299,0.94334,0.11482,0.50109,-0.55162,0.10768,-0.26433,-0.31064,-0.76526,1.0147,-0.73178,0.3323,-0.13657,-0.14013,-0.17878,0.31259,0.36723,0.55499,-0.2664,0.31862,0.60225,0.078919,0.18282,0.23014,-0.27835,0.16392,0.58131,1.052,0.062394,0.064929,-0.11918,-0.2167,-0.26018,-0.21649,-0.032523,-0.05763,-0.62775,-0.29148,0.13633,0.72715,1.535,0.08501,0.76137,0.54934,1.5163,-0.26612,0.58127,0.30339,1.8464,-0.17161,0.30309,-0.070279,0.28016,-0.036628,0.67464,0.34753,0.26499,-0.17856,-0.61162,-0.17941,-0.01017,0.21112,0.026389,0.64566,-0.39685,0.31403,-0.21637,1.1253,0.41434,0.77759,0.20421,0.70657,0.53273,-0.67276,-0.65656,0.6997,-0.06428,-0.13546,-0.15055,0.99804,1.4673,-0.47155,-0.056436,0.84609,-0.23068,-0.55114,0.68975
CATMA5A25890,-0.23072,0.82252,0.32512,0.13284,0.14406,0.005071,0.40953,-0.11252,-0.6367,-0.048073,0.21325,-0.046316,-0.35078,1.1023,-0.019882,-0.55825,-0.34621,0.1114,0.030302,-0.08142,0.19157,0.5152,-0.034952,0.20984,1.0547,0.4816,0.31078,0.14731,0.31811,-0.27275,-0.47217,-0.043207,0.29109,0.41192,-1.069,0.23357,-0.17453,0.015661,-0.26869,0.14587,0.19336,0.5926,-0.1583,0.61242,0.18635,0.01207,-0.21772,0.13986,1.9305,0.3349,0.62954,-0.04455,1.121,-0.14155,-0.33772,-0.42633,0.41828,-0.29794,-0.7107,1.4786,-0.37442,-0.45371,0.16175,-0.057764,-0.11803,-0.037838,0.25868,0.88605,0.33225,-0.24331,0.93084,-0.33266,0.9762,0.24755,0.30857,0.14661,1.8351,0.40313,0.45084,-0.52105,-0.31271,-0.37122,-0.28615,-0.0023692,-0.43072,-0.24417,-0.5184,0.13874,-0.32606,-0.1705,2.1396,-0.77657,2.3056,0.32096,0.29337,-0.034746
CATMA5A31940,4.4307,4.8735,4.7911,5.8201,5.9587,5.576,5.9084,5.849,5.2832,5.5423,5.0275,4.5254,4.5054,3.0696,5.6512,4.3986,4.9856,4.9222,4.951,4.5506,4.2185,4.3184,5.4783,4.4515,3.9397,5.6269,5.0988,6.1254,5.7446,5.6146,5.452,5.2047,5.3527,5.93,4.8931,5.5529,3.9876,4.1072,4.0583,5.4324,5.4681,4.392,4.5189,5.0629,4.9347,5.2165,4.6681,4.7428,4.1539,6.3376,5.353,6.2896,5.5844,5.1654,5.4391,5.3933,5.261,5.6974,4.5142,3.9851,3.569,4.0946,4.6662,5.1085,5.0319,4.4598,4.8672,3.7326,5.3032,5.0448,4.7771,4.2209,4.1886,4.284,5.4681,5.8331,4.7683,5.6614,5.3313,5.2017,5.1002,5.0642,4.4867,4.8576,4.1454,4.0971,5.0368,4.845,4.753,5.5496,2.9838,5.1394,3.342,4.1246,5.0121,4.618
Code proper (BHC is from Bioconductor):
library(argparse)
library(RColorBrewer)
library(BHC)
library(parallel)
library(gplots)
#data prep and clustering, to get dendrograms etc as shown in plot
data = read.csv('input.csv',header=TRUE,row.names=1,check.names=FALSE)
genes = rownames(data)
samples = colnames(data)
data = data.matrix(data)
standardisedData = (data-mean(data))/sd(data)
samples = as.numeric(samples)
hc = bhc(standardisedData, genes, timePoints=samples, dataType='time-course', verbose=TRUE)
#the plotting proper
png('heatmap.png',width=6,height=6,units='in',res=300)
heatmap.2(standardisedData, Colv=NA, Rowv=hc, tracecol=NA, scale="none", col=brewer.pal(11,'RdBu'))
#heatmap.2(standardisedData, Colv=NA, Rowv=hc, tracecol=NA, scale="none", col=brewer.pal(11,'RdBu'), margins=c(3,10)) <- this fixes it on a single-case basis
dev.off()

I would suggest just giving it a fairly wide default margin, such that it should be clipped even with longer labels, e.g.:
heatmap.2(standardisedData, Colv=NA, Rowv=hc, tracecol=NA, scale="none", col=brewer.pal(11,'RdBu'), margins=c(14, 14))
The default margin size is (5,5). The above increases the both the column and row margins, but you can increase just one or the other as well.
If you want to be more precise, you could try using max(strwidth(rownames(standardisedData))) to determine the amount of space needed by the largest label, and then convert that into the units expected by par(mar=*) (what heatmap.2() uses to specify plot margins.)
I think this should do the trick:
# determine margins to use
png('/tmp/trash', width=6,height=6,units='in',res=300)
plot.new()
margin_width = max(strwidth(rownames(standardisedData), units='inches')) * par('fin')[1]
dev.off()
#the plotting proper
png('heatmap.png',width=6,height=6,units='in',res=300)
heatmap.2(standardisedData, Colv=NA, Rowv=hc, tracecol=NA, scale="none", col=brewer.pal(11,'RdBu'), margins=c(5, margin_width))
dev.off()

How does R's tikzDevice deal with raster images?

I have read, if tikz takes a raster image it is to be stored as png. Having that, tikz produces the rest of the graph around it and include the raster image in the final tex-file back again.
Now I have the following:
pic <- T
if(pic)
{
tikz(file=paste(plotpath,"Rohdaten_S1_S2_D21.tex",sep=""),width=width,height=height,engine = "pdftex",)
#png(filename=paste(plotpath,"Rohdaten_S1_S2_D6.png",sep=""),width=width,height=height,res=res,units="in")
par(mfrow=c(2,1),mar=c(1.1,3,2,0),mgp=c(1.5,0.5,0),ps=f.size,cex=1,xaxt="n")
}
if(!pic) par(mfrow=c(2,1),mar=c(1,4,3,0))
for(i in 1:2)
{
x <- sensors[[i]]$time
y <- sensors[[i]]$depth
z <- sensors[[i]]$velo
image(x,y,z)
# plot.image(x,y,z
# ,xlim=c(max(x)-400,max(x)),zlim=2*c(-1,1)
# ,xlab="",ylab="$d/\\mathrm{m}$",zlab="$v/(\\mathrm{mm/s})$"
# ,z.adj=c(0,0),ndz=5,z.cex=1
# )
abline(v=(1:10)/0.026+par("usr")[1],lty=2)
if(!pic) abline(h=(1:floor(max(y/0.02)))*0.02)
mtext(text=paste("Sensor",i),side=3,line=0.1,adj=0)
par(mar=c(3,3,0.1,0),xaxt="s")
}
title(xlab="t/s")
if(pic) dev.off()
even the simple image() function will produce a 100MB large .tex file.
No png is produced, everything is in the .tex file?!
What am I doing wrong? Is there a switch to be set TRUE? What do I have to do to put the rasterimage apart from the nice looking text.
Thank you for your help.

The solution is quite simple, but not obvious.
the image()-function in R produces vector graphics in the first
instance. There is a switch image(...,useRaster = T) with which
one can force the image()-function to produce raster graphics.
the image()-function aspects a regular grid (quadratic pixels). Otherwise an error occurs.
How to get a regular grid?
Suppose you have an image with the coordinates x[],y[] and the scalar matrix z[,]. Then the re-sampled regular grid can be calculated:
x.new<-seq(min(xlim),max(xlim),length.out=dim.max[1])
y.new<-seq(min(ylim),max(ylim),length.out=dim.max[2])
z<-apply(z,2,function(y,x,xout) return(approx(x,y,xout=xout+min(diff(x))/2,method="constant",rule=2)$y),x,x.new)
z<-t(apply(z,1,function(y,x,xout) return(approx(x,y,xout=xout+min(diff(x))/2,method="constant",rule=2)$y),y,y.new))
tikz(file ='a.tex',width = 2, height = 2)
image(x,y,z,useRaster = T)
dev.off()
The important things are the method = "constant" and the rule = 2 statements in the approx()-function. These enables a "shifting" to the regular grid.
Applying all this and tikz() will split the picture in a a.tex-file and a a_ras1.png-file.
I hope this will help sombody programming R and using tikzDevice to produce pictures for tex documents.

fast way to read in png, add grid & coords, and output

I have a lot of png files of a floor plan (mapping) layout that I want to:
read into R
Add grid lines
Add coordinates per cell in grid
Output
There are 1000 of these files so I'm looking for a speedy method. What would be a fast way to accomplish this task. These don't need to be publication quality as I'm looking for certain behavior clusters within cells and want recording the coordinated for these events for each of the 100 frames (pngs).
Here is a MWE that produces 10 png files:
x <- y <- seq(-4*pi, 4*pi, len = 27)
r <- sqrt(outer(x^2, y^2, "+"))
dir.create("delete_me")
wd <- getwd()
setwd("delete_me")
lapply(1:10, function(x){
png(sprintf("file_%s.png", x))
image(z = z <- cos(r^2)*exp(-r/x))
dev.off()
})
setwd(wd)
The final output will look like this for each png (with all the coords filled in).
I assume grid will be the way to create the gridlines quickly but am not sure about reading the png in quickly or plotting the coordinates (assume we'll use a 10 x 10 grid on each png).

How about using ggplot() and annotation_custom() to plot the image across the entire plot area, then manually overplot the grid lines.
(In the image, I trimmed the excess whitespace and axis from the png file in advance)
# pre-req libraries
require(ggplot2)
require(grid) # rasterGrob function
require(png) # to read the PNG file
width<-10
height<-10
# generate the points and labels for the grid
points<-data.frame(expand.grid(w=1:width,h=1:height))
points$labs<-paste0("(",points$w,",",points$h,")")
points$x<-points$w-0.5 # center
points$y<-points$h-0.5
# make the gridline co-ordinates
gridx<-data.frame(x=0:width,xend=0:width,y=rep(0,width+1),yend=rep(height,width+1))
gridy<-data.frame(x=rep(0,height+1),xend=rep(width,height+1),y=0:height,yend=0:height)
grids<-rbind(gridx,gridy)
# function to plot using ggplot with annotation_custom for the image
plotgrid<-function(file){
g<-ggplot(points)+theme_bw()+
annotation_custom(rasterGrob(readPNG(file),0,0,1,1,just=c("left","bottom")),0,width,0,height)+
geom_text(aes(x=x,y=y,label=labs))+
geom_segment(aes(x=x,xend=xend,y=y,yend=yend),data=grids) +
coord_cartesian(c(0,width),c(0,height))
return(g)
}
# run the function for each file in the folder
setwd("delete_me")
lapply(list.files(),function(x)plotgrid(x))
setwd(wd)

Resources