[R] heatmap from xyz data

[Vladimir Eremeev]

You can use 

plot(y~x,col=color.index.in.palette.defined.from(z),pch=20,type="p")

where 

color.index.in.palette.defined.from(z)

is a function or an expression, returning either a color index in a
predefined palette or any other color representation, suitable for R. This
is described in ?par. 
I have successfuly used col=z*10 with your sample data.

type="p" specifies, that points will be plotted.
pch=20 says to plot them as small filled circles, sometimes I needed smaller
circles, and pch=183 worked for me.

The package fields and several other packages related to the spatial
statistics are able to plot irregularly distributed point data and
interpolate them to a regular grid (produce a 'contiguous' map).

Here are some links:

http://sal.uiuc.edu/csiss/Rgeo/
http://r-spatial.sourceforge.net/

You can also try
RSiteSearch("spatial")
or
help.search("spatial")
from the R console


Trevor Graham wrote:
> 
> I've got some data in a data frame arranged like this:
> x	y	z	othervariables	....
> 0.1	0.2	1.7	0.01	....
> 0.2	0.2	1.3	0.23	....
> 0.2	0.3	1.1	0.43	....
> etc
> 
> I'd like to plot a heatmap of this data, with x and y as the (x,y) coords
> and z defining the colour at each point.
> 
> I'm having difficulty finding the correct functions to use.  Can anyone
> make any recommendations?
> 
> It seems straightforward to make the plot using the image function if  
> the data is in matrix form, but I'm not sure how to transform my xyz  
> data into matrix form.
> 
> 

-- 
View this message in context: http://www.nabble.com/-R--heatmap-from-xyz-data-tf3180966.html#a8827597
Sent from the R help mailing list archive at Nabble.com.