gt;
>
>
> *From:* greg holly [mailto:mak.hho...@gmail.com]
> *Sent:* Monday, February 12, 2018 7:07 PM
> *To:* Richard M. Heiberger
> *Cc:* PIKAL Petr ; r-help mailing list <
> r-help@r-project.org>
>
> *Subject:* Re: [R] plotting the regression coefficients
>
>
&g
. Heiberger
Cc: PIKAL Petr ; r-help mailing list
Subject: Re: [R] plotting the regression coefficients
Hi Petr and Richard;
Thanks for your responses and supports. I just faced a different problem. I
have the following R codes and work well.
p <- ggplot(a, aes(x=Phenotypes, y=Metaboli
levels(temp$variable)
> > [1] "y1" "y2" "y3" "y4"
> >> levels(temp$variable) <- levels(temp$variable)[c(2,4,1,3)]
> >> levels(temp$variable)
> > [1] "y2" "y4" "y1" "y3"
> >>
>
gt; >> levels(temp$variable) <- levels(temp$variable)[c(2,4,1,3)]
> >> levels(temp$variable)
> > [1] "y2" "y4" "y1" "y3"
> >>
> >
> > And you will get graphs with this new levels ordering.
> >
> >
t;> levels(temp$variable)
> [1] "y2" "y4" "y1" "y3"
>>
>
> And you will get graphs with this new levels ordering.
>
> Cheers
> Petr
>
> From: greg holly [mailto:mak.hho...@gmail.com]
> Sent: Monday, February 12, 2018 8:52 AM
"y4" "y1" "y3"
>
And you will get graphs with this new levels ordering.
Cheers
Petr
From: greg holly [mailto:mak.hho...@gmail.com]
Sent: Monday, February 12, 2018 8:52 AM
To: PIKAL Petr
Cc: r-help mailing list
Subject: Re: [R] plotting the regression coefficient
t())
>
> }
>
> dev.off()
>
>
>
> But the real code partly depends on your real data.
>
>
>
> Cheers
>
> Petr
>
>
>
> *From:* greg holly [mailto:mak.hho...@gmail.com]
> *Sent:* Saturday, February 10, 2018 9:05 PM
>
> *To:* PIKAL Petr
>
10, 2018 9:05 PM
To: PIKAL Petr
Cc: r-help mailing list
Subject: Re: [R] plotting the regression coefficients
Hi Peter;
The R code you provided works very well. Once again thanks so much for this.
The number of variables in my data set that should appear on the y-axis is 733
and they are not num
reshaped
>>
>>
>>
>> library(reshape2)
>>
>> library(ggplot2)
>>
>>
>>
>> temp <- melt(temp)
>>
>> p <- ggplot(temp, aes(x=par1, y=variable, size=abs(value),
>> colour=factor(sign(value
>>
>> p+g
Hi Bert;
Thanks so much for this. It is much appreciated.
Regards,
Greg
On Thu, Feb 8, 2018 at 3:29 PM, Bert Gunter wrote:
> Fwiw, encoding magnitude in color is generally a bad idea. Using
> area(*not* radius) is also not great, but maybe it will work for you.
>
> See here for some explanatio
>
> Cheers
>
> Petr
>
> And preferably do not post in HTML, the email content could be scrambled.
>
>
>
> *From:* greg holly [mailto:mak.hho...@gmail.com]
> *Sent:* Thursday, February 8, 2018 9:23 AM
> *To:* PIKAL Petr
> *Cc:* r-help mailing list
> *Subject:*
Fwiw, encoding magnitude in color is generally a bad idea. Using area(*not*
radius) is also not great, but maybe it will work for you.
See here for some explanation:
https://www.amazon.com/Visual-Display-Quantitative-Information/dp/0961392142/ref=sr_1_1?s=books&ie=UTF8&qid=1518092778&sr=1-1&keywor
(temp)
p <- ggplot(temp, aes(x=par1, y=variable, size=abs(value),
colour=factor(sign(value
p+geom_point()
Is this what you wanted?
Cheers
Petr
And preferably do not post in HTML, the email content could be scrambled.
From: greg holly [mailto:mak.hho...@gmail.com]
Sent: Thursday, Februa
you try ggplot?
>
> Cheers
> Petr
>
>
> > -Original Message-
> > From: R-help [mailto:r-help-boun...@r-project.org] On Behalf Of greg
> holly
> > Sent: Thursday, February 8, 2018 8:14 AM
> > To: r-help mailing list
> > Subject: [R] plotting the re
gt; Subject: [R] plotting the regression coefficients
>
> Hi Dear all;
>
> I would like to create a plot for regression coefficients with each
> independent
> variable (x) along the side and the phenotypes (y) across the top (as given
> below). For each data point, direction and magn
Hi Dear all;
I would like to create a plot for regression coefficients with each
independent variable (x) along the side and the phenotypes (y) across the
top (as given below). For each data point, direction and magnitude of
effect could be color and significance could be the size of the circle? I
16 matches
Mail list logo
