[BioC] Heatmap function

[Liaw, Andy]

> From: Mark Reimers [mailto:Mark.Reimers at biosci.ki.se] 
> 
> Hi Marcus,
> 
> You can calculate your own distances, using for example sum( 
> abs(data[,i]-data[,j], na.rm=T) to compute the 'Manhattan' 
> distance between samples i & j. If you put these into a 
> symmetric matrix X, the R function as.dist(X) will transform 
> these into a distance object you can use with hclust().

But this effectively treats components containing NAs as contributing 0 to
the distance.  Is that really a reasonable thing to do?  As an example:

point #1:  3, 7, NA, 6, 1
point #2:  5, NA, 1, 8, 0

Ignoring the NAs gives dist(1, 2) = |3-5| + |6-8| + |1-0|, which is the same
as having:

point #1:  3, 7,  1, 6, 1
point #2:  5, 7,  1, 8, 0

Is that really what you want?

Andy
 
> regards
> 
> Mark
> 
> >Hmm, your answer left me thinking about how to measure 
> distances. Why 
> >doesnt a distace function just calculate the distance between the 
> >values that are there and leave out the NA:s? I have 
> filtered away with 
> >the B-test the spots that are supposedly not to be differentially 
> >expressed and have only a subset of the total number of spots. Three 
> >slides of my 18 have many NA:s. Should I exclude them 
> therefor because 
> >the distance is to affected?
> >
> >/ Marcus
> >
> >
> **************************************************************
> *****************************
> Marcus Gry Björklund
> 
> Royal Institute of Technology
> AlbaNova University Center
> Stockholm Center for Physics, Astronomy and Biotechnology 
> Department of Molecular Biotechnology 106 91 Stockholm, Sweden
> 
> Phone (office): +46 8 553 783 39
> Fax: + 46 8 553 784 81
> Visiting adress: Roslagstullsbacken 21, Floor 3
> Delivery adress: Roslagsvägen 30B
> 
> 
> 
> -- 
> Mark Reimers
> Assistant Professor, Biosciences, and
> Statistical Lead, Bioinformatics and Expression Analysis 
> Facility, Karolinska Institute Stockholm (by WebMail)
> 
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at stat.math.ethz.ch 
> https://www.stat.math.ethz.ch/mailman/listinfo> /bioconductor
>