[BioC] heatplot() 1. red-*black*-green palette, 2.suppress drawing of the row dendrogram (but NOT row reordering)

[Karl Brand]

Hi Heidi,

Thank you for your thoughts and suggestions,

On 8/7/2010 6:49 PM, Heidi Dvinge wrote:
> Hello Karl,
>
> I'm not familiar with the made4 package, but it looks like some of the
> changes you want cannot override the defaults from heatplot. Are there any
> particular reason why you don't want to use heatmap.2 directly? Using your
> example from below, then

The stated defaults of heatplot() follow the methodology by Eisen et al 
1998. This is my starting point, and i must admit, my data 'looks' v. 
good using these distance (Pearson Correlation coefficient) and linkage 
(average) methods.

*If* i knew how to implement these in heatmap.2() i would. I assume the 
arguments: "hclustfun =" and "distfun =" could be modified to deliver 
what ever i want, including Pearson Correlation coefficient and linkage 
average linkage methods. But it would be trial, error and googling to 
acheive this. Which is my next step, unless some one has the patience to 
provide an example or point me to one (cheers!). So this is my sole 
reason for using heatplot over heatmap.2 (or heatmap).

>
> library(gplots)
> my.mat<- matrix(rnorm(250), nrow=50, ncol=5)
> heatmap.2(my.mat, col=c(redgreen(256)), dendrogram="col", trace="none",
> density.info="none")
>
> will give you something similar to
>
> heatplot(my.mat, dend="both", cols.default=TRUE)
>
> only with the colours you want, and without the row dendrogram (but with
> row reordering).
>
> If you want to use heatplot() because you actually have some made4 defined
> object and not a matrix to begin with, then note that the first couple of
> lines in the heatplot() code are
>
> data<- array2ade4(dataset)
> data<- as.matrix(data)
>
> Presumably you can call something like that on your data first, to get a
> matrix that can be plotted with heatmap.2
>
> On an different note, then green-black-red isn't really the optimal colour
> scheme due to the number of red/green colour blind people. Consider using
> maybe blue-black-yellow or something else instead.

I never considered this aspect. Indeed i will.

Thanks again for your thoughts and any further ones you (or anyone else) 
may have,

Karl

The RColorBrewer
> package also has a range of pre-defined colour schemes. Check out e.g.
> display.brewer.all(9)
>
> HTH
> \Heidi
>
>> Hi Jim,
>>
>> Sorry- guilty of impatience/laziness.
>>
>> Find below what i ran on my machine and the errors i received. I thought
>> given heatplot() is based on heatmap.2(), i could just use some
>> heatmap.2() arguments to get what i wanted. I failed and lack the
>> understanding to modify heatplot() myself to achieve my goal-
>> red-BLACK-green colors and suppressing drawing of the row dendrogram.
>>
>> All tips greatly appreciated,
>>
>> Karl
>>
>> ###run local###
>> # Bioc examples heatplot() returned errors:
>> library(made4)
>> my.mat<- matrix(rnorm(250), nrow=50, ncol=5)
>>
>> # 1. default works fine:
>> windows()
>> plot.new()
>> heatplot(my.mat, dend="both", cols.default=TRUE)
>>
>> # 2. attempt  red/black/green palette with hetamap.2 col= arg
>> windows()
>> plot.new()
>> heatplot(my.mat, dend="both", cols.default=FALSE, col = c(redgreen(256)))
>>
>> # 3. attempt to supress drawing row dendrogram tree not NOT row reordering
>> windows()
>> plot.new()
>> heatplot(my.mat, dend="both", cols.default=TRUE, dendrogram = "column")
>>
>> ###end
>>
>> #2. returned:
>>   >Error in heatmap.2(data, Colv = Colv, Rowv = Rowv, col = plotcols,
>> scale = scale,  :
>>   >   formal argument "col" matched by multiple actual arguments
>>
>> #3. returned:
>>   >Error in heatmap.2(data, Colv = Colv, Rowv = Rowv, col = plotcols,
>> scale = scale,  :
>>   >   formal argument "dendrogram" matched by multiple actual arguments
>>
>>   >  sessionInfo()
>> R version 2.10.0 (2009-10-26)
>> i386-pc-mingw32
>>
>> locale:
>> [1] LC_COLLATE=English_United States.1252  LC_CTYPE=English_United
>> States.1252
>> [3] LC_MONETARY=English_United States.1252 LC_NUMERIC=C
>>
>> [5] LC_TIME=English_United States.1252
>>
>> attached base packages:
>>    [1] grid      grDevices datasets  splines   graphics  stats     tcltk
>>       utils     methods
>> [10] base
>>
>> other attached packages:
>>    [1] made4_1.20.0         scatterplot3d_0.3-29 RColorBrewer_1.0-2
>> ade4_1.4-14
>>    [5] gplots_2.7.4         caTools_1.10         bitops_1.0-4.1
>> gtools_2.6.1
>>    [9] reshape_0.8.3        plyr_0.1.9           gdata_2.7.1
>> svSocket_0.9-48
>> [13] TinnR_1.0.3          R2HTML_1.59-1        Hmisc_3.7-0
>> survival_2.35-7
>>
>> loaded via a namespace (and not attached):
>> [1] cluster_1.12.1  lattice_0.17-26 svMisc_0.9-56   tools_2.10.0
>>
>>
>>
>> On 8/6/2010 7:34 PM, James W. MacDonald wrote:
>>> Hi Karl,
>>>
>>> You need to show us the function call you used rather than saying what
>>> you did. The errors indicate that you used the 'col' argument twice in
>>> the first attempt and the 'dendrogram' argument twice in the second
>>> attempt. But without seeing exactly what you did, nobody can say for
>>> sure.
>>>
>>> Best,
>>>
>>> Jim
>>>
>>>
>>>
>>> On 8/6/10 11:55 AM, Karl Brand wrote:
>>>> Esteemed BioC Users,
>>>>
>>>> Could some one help me obtain the garden variety microarray heatmap
>>>> "red-*black*-green" palette in heatplot()? Using-
>>>>
>>>> lowcol="green", highcol="red"
>>>>
>>>> -has no black transition between red and green.
>>>>
>>>> Since its based on heatmap.2, i thought i could achieve this by
>>>> passing the following argument to heatplot():
>>>>
>>>> cols.default=FASLE #to suppress default color palette
>>>> col = c(redgreen(256))
>>>>
>>>> but get the error-
>>>>
>>>>> Error in heatmap.2(data, Colv = Colv, Rowv = Rowv, col = plotcols,
>>>> scale = scale, :
>>>>> formal argument "col" matched by multiple actual arguments
>>>>
>>>> I'd also like to suppress drawing of the row dendrogram (but NOT the
>>>> reordering or rows), but again the heatmap.2 argument to do this-
>>>>
>>>> dendrogram = "column" #draw only column dendrograms
>>>>
>>>> returned-
>>>>
>>>>> Error in heatmap.2(data, Colv = Colv, Rowv = Rowv, col = plotcols,
>>>> scale = scale, :
>>>>> formal argument "dendrogram" matched by multiple actual arguments
>>>>
>>>> With thanks in advance, cheers,
>>>>
>>>> Karl
>>>>
>>>
>>
>> --
>> Karl Brand
>> Department of Genetics
>> Erasmus MC
>> Dr Molewaterplein 50
>> 3015 GE Rotterdam
>> T +31 (0)10 704 3457 |F +31 (0)10 704 4743 |M +31 (0)642 777 268
>>
>> _______________________________________________
>> Bioconductor mailing list
>> Bioconductor at stat.math.ethz.ch
>> https://stat.ethz.ch/mailman/listinfo/bioconductor
>> Search the archives:
>> http://news.gmane.org/gmane.science.biology.informatics.conductor
>>
>
>

-- 
Karl Brand
Department of Genetics
Erasmus MC
Dr Molewaterplein 50
3015 GE Rotterdam
T +31 (0)10 704 3457 |F +31 (0)10 704 4743 |M +31 (0)642 777 268