Luigi I suspect that it is in accessing the correct groups within the panel function
I suggest that printing the values before plotting within the panel function may give you some idea. Without specific data I cannot suggest anything else Duncan -----Original Message----- From: Luigi Marongiu [mailto:marongiu.lu...@gmail.com] Sent: Wednesday, 7 May 2014 22:37 To: Duncan Mackay Subject: Re: [R] set axis limit in lattice Dear Duncan, thank you for the reply. I have fixed the axis, bu still I don't understand why i cannot draw the segments on my real data whereas i could on the example when the panel function is quite general. igra is the outcome that i rendered as positivity in the example. i also change it into numeric values (1/0) to resemble more closely the example but still the segments are ignored. may i ask you for a tip on the possible causes? best regards luigi On Thu, May 1, 2014 at 3:00 AM, Duncan Mackay <dulca...@bigpond.com> wrote: > If you just want to limit the axis just insert as arguments > for the x axis > xlim = c(min, max), > and for the y > ylim = c(min,max) , > where min and max correspond to the appropriate min and max for the x and y > axes > > As for the panel.groups you mention medians where I was using means - if you > want medians you will have to amend means to medians > > Make sure for the groups you have the right mode numeric/ factor ? > I do not know what igra is > > Duncan > > -----Original Message----- > From: Luigi Marongiu [mailto:marongiu.lu...@gmail.com] > Sent: Thursday, 1 May 2014 03:54 > To: Duncan Mackay > Subject: Re: [R] set axis limit in lattice > > Dear Duncan, > these solutions are actually splitting the drawing in different panels > and the scale=free is giving different limits to each panel. I still > need the relation=same but just to input myself the fixed limits of > the axis. It should go into the scale argument but i don't know the > actual syntax. > As for the segments, you were perfectly right, there was a problem > with my script so the medians did not match the actual data. Your > solution using panel.groups works on the example, but when I copied it > to the actual dataframe I am working with it did not work despite the > fact of being of quite a general syntax. could you please give a look > at the dataframe I have attached and see where I made a mistake in > applying your solution? what I wrote is: > > my.data.IP10<-subset(my.data, gene=="i" & stimulus != "Unstimulated") > > stripplot( > my.data.IP10$ratio ~ > factor(my.data.IP10$igra)|factor(my.data.IP10$stimulus, levels = > c("ESAT6","CFP10","Rv3615c", "Rv2654","Rv3879", "Rv3873","PHA")), > my.data, > group = my.data.IP10$igra, > hor=F, > layout = c(7,1), > scales = list(x = list(draw=FALSE)), > jitter.data=TRUE, > pch=c(1,16), > col="black", > ylab=expression(bold("Ratio GOI/RG")), > xlab=expression(bold("Antigen stimulation")), > main="IP10 expression", > par.settings = list(strip.background=list(col="light grey")), > par.strip.text=list(font=2), > key = list(space="top", columns=2, text=list(c("IGRA Negative", > "IGRA Positive"), col="black"), points=list(pch=c(1,16), > col="black")), > > panel = panel.superpose, > panel.groups = function(x, y, group.number, ...) { > > panel.segments(x0= x-0.25, x1=x+0.25, > y0=mean(y), y1=mean(y), lty = c(1:2)[group.number], > lwd=c(2:1)[group.number], col = c(1:2)[group.number]) > > panel.stripplot(x,y, ...) > } > ) > > > ### Many thanks > ### Best regards > Luigi > > > On Wed, Apr 30, 2014 at 12:15 AM, Duncan Mackay <dulca...@bigpond.com> wrote: >> Hi Luigi >> >> This produces plenty of white space for the y axes >> >> Y<-max(my.data$copy) >> >> stripplot( >> copy ~ factor(positivity)|factor(stimulation, levels = c("Unst.", >> "ESAT6","CFP10","Rv3615c", "Rv2654","Rv3879", "Rv3873","PHA")), >> my.data, >> group = positivity, >> hor=F, >> layout = c(8,1), >> scales = list(x = list(at = c(1,2), labels = c("N","P")), >> y= list(relation = "free")), >> jitter.data=TRUE, pch=c(16,1), col="black", >> ylab=expression(bold("Copy")), >> xlab=expression(bold("Stimulation")), >> main="Plot", >> par.settings = list(strip.background=list(col="white")), >> par.strip.text=list(font=2), >> key = key.plot, >> panel = function(x, y, ...) >> { >> pnl = panel.number() >> k<-0 >> for (i in 1:2){ >> datme<-subset(datmeA, positivity==i-1) >> w<-i+k >> panel.segments(x0 = X[w], y0 = datme[datme[,2]== >> levels(datme[,2])[pnl],"copy"], >> x1 = X[w+1], y1 = datme[datme[,2]== >> levels(datme[,2])[pnl],"copy"],lwd = 2, col = "black") >> k=k+1 >> >> panel.stripplot(x,y, ...) >> } >> } >> ) >> You could limit the y-axis to your levels >> stripplot( >> copy ~ factor(positivity)|factor(stimulation, levels = c("Unst.", >> "ESAT6","CFP10","Rv3615c", "Rv2654","Rv3879", "Rv3873","PHA")), >> my.data, >> group = positivity, >> hor=F, >> layout = c(8,1), >> scales = list(x = list(at = c(1,2), labels = c("N","P")), >> y= list(relation = "free", >> limits = list(c(0, max panel 1), >> c(0, max panel 2), >> ... >> c(0, max panel 8)), >> )), >> jitter.data=TRUE, pch=c(16,1), col="black", >> ylab=expression(bold("Copy")), >> xlab=expression(bold("Stimulation")), >> main="Plot", >> par.settings = list(strip.background=list(col="white")), >> par.strip.text=list(font=2), >> key = key.plot, >> panel = function(x, y, ...) >> { >> pnl = panel.number() >> k<-0 >> for (i in 1:2){ >> datme<-subset(datmeA, positivity==i-1) >> w<-i+k >> panel.segments(x0 = X[w], y0 = datme[datme[,2]== >> levels(datme[,2])[pnl],"copy"], >> x1 = X[w+1], y1 = datme[datme[,2]== >> levels(datme[,2])[pnl],"copy"],lwd = 2, col = "black") >> k=k+1 >> >> panel.stripplot(x,y, ...) >> } >> } >> ) >> >> Sorting the stimulation groups shows that some of them can be loosely >> grouped so reordering the factors >> >> stripplot(copy ~ factor(positivity)|factor(stimulation, levels = >> c("PHA","Rv2654","CFP10","Unst.","Rv3615c", "Rv3873","ESAT6", "Rv3879")), >> my.data, >> group = positivity, >> hor=F, >> layout = c(2,4), >> scales = list(x = list(at = c(1,2), labels = c("N","P")), >> y= list(relation = "free")), >> jitter.data=TRUE, pch=c(16,1), col="black", >> ylab=expression(bold("Copy")), >> xlab=expression(bold("Stimulation")), >> main="Plot", >> par.settings = list(strip.background=list(col="white")), >> par.strip.text=list(font=2), >> key = key.plot, >> panel = function(x, y, ...) >> { >> pnl = panel.number() >> k<-0 >> for (i in 1:2){ >> datme<-subset(datmeA, positivity==i-1) >> w<-i+k >> panel.segments(x0 = X[w], y0 = datme[datme[,2]== >> levels(datme[,2])[pnl],"copy"], >> x1 = X[w+1], y1 = datme[datme[,2]== >> levels(datme[,2])[pnl],"copy"],lwd = 2, col = "black") >> k=k+1 >> >> panel.stripplot(x,y, ...) >> } >> } >> ) >> >> or using panel.groups >> >> stripplot(copy ~ factor(positivity)|factor(stimulation, levels = >> c("PHA","Rv2654","CFP10","Unst.","Rv3615c", "Rv3873","ESAT6", "Rv3879")), >> my.data, >> group = positivity, >> hor=F, >> layout = c(2,4), >> scales = list(x = list(at = c(1,2), labels = c("N","P")), >> y= list(relation = "free")), >> jitter.data=TRUE, pch=c(16,1), col="black", >> ylab=expression(bold("Copy")), >> xlab=expression(bold("Stimulation")), >> main="Plot", >> par.settings = list(strip.background=list(col="white")), >> par.strip.text=list(font=2), >> key = key.plot, >> panel = panel.superpose, >> panel.groups = function(x, y, group.number, ...) { >> >> panel.segments(x0= x-0.25, x1=x+0.25, y0=mean(y), >> y1=mean(y), lty = c(1:2)[group.number], >> lwd=c(2:1)[group.number], col = c(1:2)[group.number]) >> >> panel.stripplot(x,y, ...) >> } >> ) >> >> Regards >> >> Duncan Mackay >> >> -----Original Message----- >> From: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] On >> Behalf Of Luigi Marongiu >> Sent: Wednesday, 30 April 2014 03:58 >> To: r-help@r-project.org >> Subject: [R] set axis limit in lattice >> >> Dear all, >> >> I would like to set the axis of a figure using the max() so to have >> more control on the limits of the axis -- this because in the actual >> case more independent figures are generated from the same dataframe >> and these must have the same axis scale. >> >> Since the figure is generated using lattice device, the syntax is too >> complicated for me: when I place the xlim=c(0,Y) in the arguments — >> where Y is obtained using max(data) — there is no result. In basic R >> this argument would work virtually in any position. >> >> Any tip? >> >> Best regards >> >> Luigi >> >> ####CODE >> ### open plot library >> library(lattice) >> my.data<-structure(list( >> column_1 = 1:120, >> column_2 = structure(c( >> 1,2,3,4,5,6,7,8, >> 1,2,3,4,5,6,7,8, >> 1,2,3,4,5,6,7,8, >> 1,2,3,4,5,6,7,8, >> 1,2,3,4,5,6,7,8, >> 1,2,3,4,5,6,7,8, >> 1,2,3,4,5,6,7,8, >> 1,2,3,4,5,6,7,8, >> 1,2,3,4,5,6,7,8, >> 1,2,3,4,5,6,7,8, >> 1,2,3,4,5,6,7,8, >> 1,2,3,4,5,6,7,8, >> 1,2,3,4,5,6,7,8, >> 1,2,3,4,5,6,7,8, >> 1,2,3,4,5,6,7,8), .Label = c("Unst.", "ESAT6", "CFP10", "Rv3615c", >> "Rv2654", "Rv3879", "Rv3873", "PHA"), class = "factor"), >> column_3 = structure(c( >> 0,0,0,0,0,0,0,0, >> 0,0,0,0,0,0,0,0, >> 0,0,0,0,0,0,0,0, >> 0,0,0,0,0,0,0,0, >> 0,0,0,0,0,0,0,0, >> 0,0,0,0,0,0,0,0, >> 0,0,0,0,0,0,0,0, >> 1,1,1,1,1,1,1,1, >> 1,1,1,1,1,1,1,1, >> 1,1,1,1,1,1,1,1, >> 1,1,1,1,1,1,1,1, >> 1,1,1,1,1,1,1,1, >> 1,1,1,1,1,1,1,1, >> 1,1,1,1,1,1,1,1, >> 0,0,0,0,0,0,0,0)), >> column_4 = c( >> >> 192.0519108,183.6403531,53.46798757,83.60638077,69.60749873,159.4706861,256.8765622,499.2899303, >> >> 2170.799076,1411.349719,2759.472348,2098.973397,2164.739515,1288.676574,1611.486543,6205.229575, >> >> 870.7424981,465.9967135,191.8962375,864.0937485,2962.693675,1289.259137,2418.651212,7345.712517, >> 0,168.1198893,674.4342961,101.1575401,47.81596237,0,0,1420.793922, >> >> 142.6871331,5.466468742,291.9564635,80.73914133,73.02239621,64.47806871,144.3543635,3167.959757, >> >> 3164.748333,1092.634557,28733.20269,1207.87783,729.6090973,151.8706088,241.2466141,9600.963594, >> >> 1411.718287,12569.96285,1143.254476,6317.378481,16542.27718,79.68025792,1958.495138,7224.503437, >> >> 208.4382941,69.48609769,656.691151,0.499017582,7114.910926,187.6296174,41.73980805,8930.784541, >> >> 4.276752185,0.432300363,60.89228665,1.103924786,0.490686366,1.812993239,7.264531581,1518.610307, >> >> 2172.051528,595.8513744,17141.84336,589.6565971,1340.287628,117.350942,593.7034054,24043.61463, >> >> 0,81.83292179,1539.864321,36.41722958,8.385131047,161.7647376,65.21615696,7265.573875, >> >> 97.84753179,154.051827,0.613835842,10.06138851,45.04879285,176.8284258,18795.75462,30676.769, >> >> 5780.34957,944.2200834,2398.235596,1083.393165,2541.714557,1251.670895,1547.178549,1792.679176, >> >> 3067.988416,8117.210173,23676.02226,8251.937547,17360.80494,18563.61561,16941.865,31453.96708, >> >> 2767.493803,4796.33016,12292.93705,3864.657567,9380.673835,14886.44683,8457.88646,26050.47191)), >> .Names = c("row", "stimulation", "positivity", "copy"), row.names = >> c(NA, -120L), >> class = "data.frame") >> >> key.plot<-list( >> space="top", columns=2, >> text=list(c("Positive", "Negative"), col="black"), >> points=list(pch=c(16,1), col="black")) >> >> datmeA <- aggregate(copy ~ positivity+stimulation, my.data, median, na.rm = >> T) >> >> X<-c(0.7, 1.3, 1.7, 2.3) >> >> Y<-max(copy) >> >> stripplot( >> copy ~ factor(positivity)|factor(stimulation, levels = c("Unst.", >> "ESAT6","CFP10","Rv3615c", "Rv2654","Rv3879", "Rv3873","PHA")), >> my.data, >> group = positivity, >> hor=F, >> layout = c(8,1), >> scales = list(x = list(at = c(1,2), labels = c("N","P"))), >> jitter.data=TRUE, pch=c(16,1), col="black", >> ylab=expression(bold("Copy")), >> xlab=expression(bold("Stimulation")), >> main="Plot", >> par.settings = list(strip.background=list(col="white")), >> par.strip.text=list(font=2), >> key = key.plot, >> panel = function(x, y, ...) >> { >> pnl = panel.number() >> k<-0 >> for (i in 1:2){ >> datme<-subset(datmeA, positivity==i-1) >> w<-i+k >> panel.segments(x0 = X[w], y0 = datme[datme[,2]== >> levels(datme[,2])[pnl],"copy"], >> x1 = X[w+1], y1 = datme[datme[,2]== >> levels(datme[,2])[pnl],"copy"],lwd = 2, col = "black") >> k=k+1 >> >> panel.stripplot(x,y, ...) >> } >> } >> ) >> >> ______________________________________________ >> R-help@r-project.org mailing list >> https://stat.ethz.ch/mailman/listinfo/r-help >> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html >> and provide commented, minimal, self-contained, reproducible code. >> > ______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
