[R] Calculating the Distance

Keizer_71 christophe.lo at gmail.com
Thu Feb 21 00:18:30 CET 2008


***********creating matrix and calculating variance across probesets********

x<-1:20000

y<-2:141

data.matrix<-data.matrix(data[,y])

variableprobe<-apply(data.matrix[x,],1,var)

hist(variableprobe)

**************filter out low variance*************

data.sub = data.matrix[order(variableprobe,decreasing=TRUE),][1:10000,]

dim(data.sub)

[1] 10000   140

summary(data.sub)

a few samples:

  Sample_68_C      Sample_69_D      Sample_69_C      Sample_70_D     
Sample_70_C    
 Min.   : 1.873   Min.   : 1.893   Min.   : 1.873   Min.   : 1.722   Min.  
: 1.871  
 1st Qu.: 5.202   1st Qu.: 5.176   1st Qu.: 4.176   1st Qu.: 4.763   1st
Qu.: 5.366  
 Median : 6.559   Median : 6.502   Median : 5.579   Median : 6.208   Median
: 6.622  
 Mean   : 6.473   Mean   : 6.445   Mean   : 5.697   Mean   : 6.189   Mean  
: 6.558  
 3rd Qu.: 7.738   3rd Qu.: 7.742   3rd Qu.: 6.967   3rd Qu.: 7.547   3rd
Qu.: 7.813  
 Max.   :14.953   Max.   :14.863   Max.   :14.741   Max.   :15.102   Max.  
:14.975  

What is the best way to give me me probes only. I am trying to tell R to
show me all the probes (10,000).

What i want to do is to use the dist function to compute distances between
the samples above. This function will take the matrix and computes the
distances between the rows of the matrix.

I tried dis <- dist(t(exprs(data.sub)), method="euclidean") but it is
measuring the point by point which is too big. I would like to measure the
distances between the rows.

thanks!!!






-- 
View this message in context: http://www.nabble.com/Calculating-the-Distance-tp15601307p15601307.html
Sent from the R help mailing list archive at Nabble.com.



More information about the R-help mailing list