[R] how to create density plot in R with p value

[Ana Marija]

Hi Jim


I tried your code and it didn't show me density curves.

However I was able to do teh plot myself via:

densityplot(~dat,data=tot,
       groups=type,
       par.settings = list(superpose.line = list(col = c("blue","red"))),
       xlab="log2 (variance)",
       plot.points=FALSE,
       auto.key=TRUE)


and it is attached. The only thing I need to do is to make a line in
between peaks of the curves and put the star above that line. Not a p
value but just a star symbol.

Please advise,
Thanks
Ana

On Tue, Feb 11, 2020 at 3:23 PM Jim Lemon <drjimlemon using gmail.com> wrote:
>
> Hi Ana,
> Your image didn't make it through (as usual). Try PNG or PDF format.
>
> Assumptions
> 1) You want to plot two lines using the _paired_ values of inter- and
> intra-individual variance.
> 2) The cases are in the same order in your data
> 3) You want to identify the two lines
> 4) You want to place text with some p-value between the lines
>
> This is really (and suspiciously) simple. Here's a start:
>
> tot<-read.table(text="dat type
>  -3.962 inter.individual.variance
>  -4.301 inter.individual.variance
>  -1.690 inter.individual.variance
>  -0.375 inter.individual.variance
>   1.816 inter.individual.variance
>   0.138 inter.individual.variance
>  -4.09 intra.individual.variance
>  -4.64 intra.individual.variance
>  -5.57 intra.individual.variance
>  -2.96 intra.individual.variance
>  -4.43 intra.individual.variance
>  -3.60 intra.individual.variance",
>  header=TRUE)
> # get the numeric values of the factor 'type'
> tot$ntype<-as.numeric(tot$type)
> plot(tot$dat[tot$ntype==1],type="l",col="green",ylim=range(tot$dat))
> lines(tot$dat[tot$ntype==2],col="blue")
> text(5,-1.5,"p = 0.005")
> legend(1,1,c("Inter-individual variance","Intra-individual variance"),
>  lty=1,col=c("green","blue"))
>
> Jim
>
> On Wed, Feb 12, 2020 at 7:07 AM Ana Marija <sokovic.anamarija using gmail.com> wrote:
> >
> > so I transformed my data from the previous email to look like this:
> >
> > > head(tot)
> >      dat                      type
> > 1 -3.962 inter.individual.variance
> > 2 -4.301 inter.individual.variance
> > 3 -1.690 inter.individual.variance
> > 4 -0.375 inter.individual.variance
> > 5  1.816 inter.individual.variance
> > 6  0.138 inter.individual.variance
> > > tail(tot)
> >         dat                      type
> > 31177 -4.09 intra.individual.variance
> > 31178 -4.64 intra.individual.variance
> > 31179 -5.57 intra.individual.variance
> > 31180 -2.96 intra.individual.variance
> > 31181 -4.43 intra.individual.variance
> > 31182 -3.60 intra.individual.variance
> >
> > then I can make plot using this:
> >
> > densityplot(~dat,data=tot,
> >        groups=type,
> >        par.settings = list(superpose.line = list(col = c("blue","red"))),
> >        xlab="log2 (variance)",
> >        plot.points=FALSE,
> >        auto.key=TRUE)
> >
> > and calculate my p value with t test, say it is 0.005
> >
> > But how to add that p value in between distribution of the curves?
> >
> > Thanks
> > Ana
> >
> > On Tue, Feb 11, 2020 at 12:54 PM Ana Marija <sokovic.anamarija using gmail.com> wrote:
> > >
> > > Hi,
> > >
> > > I have data like this:
> > >
> > > > head(a)
> > >                    X   geneID inter.individual.variance
> > > intra.individual.variance F.value  p.value     CV
> > > 1 3iUZ.47hLFki49VR4o   MLLT10                    0.0642
> > >    0.01395    4.60 1.00e-05 0.0222
> > > 2 fEn1QlU0MCVe9GeR64 C1orf123                    0.0507
> > >    0.00671    7.57 1.00e-08 0.0172
> > > 3 ud_tTlU5LtB478JxIk    FSD1L                    0.3100
> > >    0.02682   11.56 1.00e-11 0.0639
> > > 4 3KV3OJIIRuJ5KJ6VkI  TXNDC11                    0.7710
> > >    0.02813   27.41 9.99e-19 0.0688
> > > 5 o_rupcEAKnQqnoh6ec   CYB5R2                    3.5209
> > >    0.03391  103.83 9.99e-31 0.1357
> > > 6 Nk_t6ULcQ7VDNChBRU     GBP1                    1.1005
> > >    0.09522   11.56 9.98e-12 0.0773
> > >
> > > I would like to create density plot, like the attached between
> > > "inter.individual.variance" and "intra.individual.variance" and have p
> > > value between those two curves.
> > >
> > > Please advise,
> > > Ana
> >
> > ______________________________________________
> > R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
> > 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.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: newp.png
Type: image/png
Size: 141973 bytes
Desc: not available
URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20200211/bd030a1f/attachment.png>