Hi Ana,
I think this is what you want in the panel style of plot. Let me know
if not, or if I have calculated the wrong percentages. The overlaid
histograms definitely use a different calculation.
amsdf<-read.table("pheno_m1_plot",header=TRUE,stringsAsFactors=FALSE)
dim(amsdf)
# find the right breaks for your "cut"
casen<-table(cut(amsdf$HBA1C[amsdf$pheno==2],breaks=3:14))
controln<-table(cut(amsdf$HBA1C[amsdf$pheno==1],breaks=3:14))
# save yourself some typing
HBA1C2<-amsdf$HBA1C[amsdf$pheno==2]
HBA1C1<-amsdf$HBA1C[amsdf$pheno==1]
ncases<-length(HBA1C2)
ncontrols<-length(HBA1C1)
split.screen(matrix(c(0,1,0.6,1,0,1,0,0.6),nrow=2,byrow=TRUE))
par(mar=c(0,4,1,2))
barpos=barplot(100*casen/ncases,names.arg=NA,col="orange",
space=0,ylab="Percentage",xaxt="n",ylim=c(0,27))
case_text<-sprintf(
"Cases: n=%d, nulls=%d, median=%.1f, mean=%.1f, sd=%.1f",
length(HBA1C2),sum(is.na(HBA1C2)),round(median(HBA1C2,na.rm=TRUE),1),
round(mean(HBA1C2,na.rm=TRUE),1),round(sd(HBA1C2,na.rm=TRUE),1))
text(mean(barpos),25,case_text)
box()
screen(2)
par(mar=c(4,4,0,2))
barplot(100*controln/ncontrols,names.arg=NA,
space=0,ylab="Percentage",col="orange",ylim=c(0,34))
control_text<-sprintf(
"Cases: n=%d, nulls=%d, median=%.1f, mean=%.1f, sd=%.1f",
length(HBA1C1),sum(is.na(HBA1C1)),round(median(HBA1C1,na.rm=TRUE),1),
round(mean(HBA1C1,na.rm=TRUE),1),round(sd(HBA1C1,na.rm=TRUE),1))
text(mean(barpos),32,control_text)
box()
library(plotrix)
staxlab(1,at=barpos,labels=names(casen))
Jim
On Sat, May 23, 2020 at 9:01 AM Ana Marija <sokovic.anamarija at gmail.com> wrote:
Hi Jim, My data is attached. It is most kind of you for looking into this! Cheers, Ana On Fri, May 22, 2020 at 5:49 PM Jim Lemon <drjimlemon at gmail.com> wrote:
Hi Ana, As I had very little idea what your data looked like, what I made up obviously didn't fit in the plot that well. If you can send the data I can make a better attempt. The other thing is whether you want a plot with two adjacent panels (what I sent) or overlaid histograms (what Eric sent). Let me know. Jim On Sat, May 23, 2020 at 12:45 AM Ana Marija <sokovic.anamarija at gmail.com> wrote:
HI Jim, Thank you so much for getting back to me I tried your codes and I got this in attach, I think the issue is in calculating percentage per groups (cases or controls) ... I can send you the whole dataset if you would like to try with it On Thu, May 21, 2020 at 11:14 PM Jim Lemon <drjimlemon at gmail.com> wrote:
-------------- next part -------------- A non-text attachment was scrubbed... Name: ams1.png Type: image/png Size: 18251 bytes Desc: not available URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20200523/ee75ea08/attachment.png>