[BioC] Colourful way of visualising differential analysis results

[Yannick Wurm]

Hi Dan,
my apologies for the slow reply.
How many different PC3M samples were used? Only 1? So you want a  
single-column heatmap (with xxxx rows, 1 per spot?)

Then just make a limma model
	then use makeContrasts(PC3M-Knockdown, design)
	then once you've done lmFit,
	you'll find the relative expression values in the "coefficients" part  
of myFitObject. Access it via myFitObject$coefficients


Alternatively, if you want one heatmap column per PC3M you hybridized,  
you can calculate them by hand:
	+go to your MA object, and use the log2 relative expression levels  
from MA$M (you have one column per microarray).
	+To compare Knockdown against PC3M,
		- use array 1_2 data as is:
			MA$M[,2]
		- calculate Knockdown-PC3M a second time via arrays 1_3 and 3_4:
			MA$M[,10]+MA$M[,3]
		- calculate Knockdown-PC3M again via arrays 2_2  and 1_1:
			-MA$M[,1] -MA$M[,5]
		- and then again via arrays 3_1 and 3_2


heatmap(cbind(MA$M[,2], MA$M[,10]+MA$M[,3], -MA$M[,1] -MA$M[,5], ...))


Best,
yannick

On Nov 11, 2008, at 11:24 AM, Daniel Brewer wrote:

> Hi,
>
> That sounds great.  I am not sure exactly how you can do it and  
> whether
> it is applicable to the experiment.  Could you provide a simple  
> example?
>
> The experiment information is below and I am interested in the PC3M vs
> knockdown comparison
>
> Targets file:
> SlideNumber	ArrayNumber	FileName	Name	Cy3	Cy5
> 1	1	Input/1_1.txt	1_1	Scramble	Knockdown
> 1	2	Input/1_2.txt	1_2	Knockdown	PC3M
> 1	3	Input/1_3.txt	1_3	PNT2	PC3M
> 1	4	Input/1_4.txt	1_4	Pooled	PNT2
> 2	2	Input/2_2.txt	2_2	PC3M	Scramble
> 2	3	Input/2_3.txt	2_3	PNT2	Scramble
> 3	1	Input/3_1.txt	3_1	PC3M	Pooled
> 3	2	Input/3_2.txt	3_2	Pooled	Knockdown
> 3	3	Input/3_3.txt	3_3	Scramble	Pooled
> 3	4	Input/3_4.txt	3_4	Knockdown	PNT2
>
> PC3M = the control cell line
> Knockdown = PC3M with an siRNA knockdown vector
> Scramble = PC3M with a vector with a scrambled sequence
> PNT2 = Another cell line (not of interest here)
> Pooled = poll of knockdowns before you get specific clone,  
> intermediate
> between PCM3 and knockdown - a hetrogenious group (not considered  
> here)
>
>> design
>      Knockdown PNT2 Pooled Scramble
> [1,]         1    0      0       -1
> [2,]        -1    0      0        0
> [3,]         0   -1      0        0
> [4,]         0    1     -1        0
> [5,]         0    0      0        1
> [6,]         0   -1      0        1
> [7,]         0    0      1        0
> [8,]         1    0     -1        0
> [9,]         0    0      1       -1
> [10,]        -1    1      0        0
>
> Thanks Dan
>
> Yannick Wurm wrote:
>> Hi Dan,
>>
>> for this kind of thing, I'll fit another limma model just to obtain
>> estimates of what needs to be visualized...
>> In one case, I needed to separately visualize expression levels from
>> each biological replicate, but variability was such that I had  
>> grouped
>> them together in my model. To estimate expression levels for each
>> biological replicate, I recreated a targets file, separating each
>> biological replicate by name. Then calculated a fit, and asked for
>> contrasts between each sample and one RNA which I chose as reference.
>> (centering expression levels within each gene afterwards works too)
>>
>> Despite a complex design it was thus possible to generate a heatmap
>> where each of the 8 biological replicated RNAs from 3 different
>> conditions where represented separately.
>>
>> hope this helps,
>>
>> yannick
>>
>>
>>
>> On Nov 10, 2008, at 17:33 , Daniel Brewer wrote:
>>
>>> Dear all,
>>>
>>> I am doing some work on a two-colour microarray (Agilent)  
>>> experiment and
>>> I have used limma to do some differential analysis.  The person I am
>>> doing this work was keen to have a heatmap of the differentially
>>> expressed genes expression levels.  Unfortunately, the design is  
>>> rather
>>> complex and random (closer to a loop design than a common  
>>> reference) so
>>> its not possible to produce a traditional heatmap.  I was  
>>> wondering if
>>> anyone had any suggestions of a colourful way to show that the
>>> expression of the two groups are different?
>>>
>>> In particular I was thinking that there must be estimates of the
>>> expression and error in each group by the linear model, but couldn't
>>> work out how to find these.
>>>
>>> Thanks
>>>
>>> Dan
>
>
>
> -- 
> **************************************************************
>
> Daniel Brewer
>
> Institute of Cancer Research
> Molecular Carcinogenesis
> MUCRC
> 15 Cotswold Road
> Sutton, Surrey SM2 5NG
> United Kingdom
>
> Tel: +44 (0) 20 8722 4109
> Fax: +44 (0) 20 8722 4141
>
> Email: daniel.brewer at icr.ac.uk
>
> **************************************************************
>
> The Institute of Cancer Research: Royal Cancer Hospital, a  
> charitable Company Limited by Guarantee, Registered in England under  
> Company No. 534147 with its Registered Office at 123 Old Brompton  
> Road, London SW7 3RP.
>
> This e-mail message is confidential and for use by the...{{dropped:13}}