R help - Jun 2005 - Crosstabs in R

HI,

	I have the data in the following format. My aim is to determine the concordance
in genotype calls (SNP1, 2, 3,etc) at two centers.
	 
	DNA center snp1 snp2
	NA07019 1 A A
	NA07348 1 M G
	NA10830 1 A G
	NA10851 1 M G
	NA10857 1 A G
	NA10860 1 A G
	NA10861 1 A G
	NA12761 1 ? R
	NA07019 2 A A
	NA07348 2 M G
	NA10830 2 A G
	NA10851 2 M G
	NA10857 2 A G
	NA10860 2 A G
	NA10861 2 A G
	NA12761 2 A A
	 
	In SPSS, I would create individual crosstabulation tables and any positive
integer values (counts) falling outside of the diagonal are disagreements in
calls by the 2 centers (SNP2).

	At SNP1, there are total of 7 genotype counts to compare (one DNA did not have
genotype readout). And 7/7 are in agreement.
	 
	At SNP2, all 8 DNA samples were successfully typed by both centers. There is
one discrepant call between centers.

	Q1: how do I create the same crosstabs in R?
	 
	Eventually, I would summarize the data as follows:
	 
	SNP N_to_compare N_agreement %Concordance
	snp1 7  7  100
	snp2 8  7  87.5
	etc
	 
	Q2: From the individual crosstabs tables, can I automatically extract those
numbers in columns 2 and 3?
	 
	 
	rgds,
	jenny

Here is one way to do it.
>
jen2<-reshape(jen,idvar="DNA",timevar="center",direction="wide")
This makes it easier to work with.
> jen2<-replace(jen2,jen2=="?",NA)
Change the question marks to NA.
> jen2$match1<-((jen2[,2]==jen2[,4])*1)
> jen2$match2<-((jen2[,3]==jen2[,5])*1)
> jen2
      DNA snp1.1 snp2.1 snp1.2 snp2.2 match1 match2
1 NA07019      A      A      A      A      1      1
2 NA07348      M      G      M      G      1      1
3 NA10830      A      G      A      G      1      1
4 NA10851      M      G      M      G      1      1
5 NA10857      A      G      A      G      1      1
6 NA10860      A      G      A      G      1      1
7 NA10861      A      G      A      G      1      1
8 NA12761   <NA>      R      A      A     NA     
0
> mean(jen2[,6:7])
match1 match2 
    NA  0.875 
> mean(jen2[,6:7],na.rm=T)
match1 match2 
 1.000  0.875 
>
-----Original Message-----
From: r-help-bounces at stat.math.ethz.ch
[mailto:r-help-bounces at stat.math.ethz.ch] On Behalf Of Tan Hui Hui Jenny
Sent: June 27, 2005 9:13 PM
To: r-help at stat.math.ethz.ch
Subject: [R] Crosstabs in R

HI,

	I have the data in the following format. My aim is to determine the
concordance in genotype calls (SNP1, 2, 3,etc) at two centers.
	 
	DNA center snp1 snp2
	NA07019 1 A A
	NA07348 1 M G
	NA10830 1 A G
	NA10851 1 M G
	NA10857 1 A G
	NA10860 1 A G
	NA10861 1 A G
	NA12761 1 ? R
	NA07019 2 A A
	NA07348 2 M G
	NA10830 2 A G
	NA10851 2 M G
	NA10857 2 A G
	NA10860 2 A G
	NA10861 2 A G
	NA12761 2 A A
	 
	In SPSS, I would create individual crosstabulation tables and any
positive integer values (counts) falling outside of the diagonal are
disagreements in calls by the 2 centers (SNP2).

	At SNP1, there are total of 7 genotype counts to compare (one DNA
did not have genotype readout). And 7/7 are in agreement.
	 
	At SNP2, all 8 DNA samples were successfully typed by both centers.
There is one discrepant call between centers.

	Q1: how do I create the same crosstabs in R?
	 
	Eventually, I would summarize the data as follows:
	 
	SNP N_to_compare N_agreement %Concordance
	snp1 7  7  100
	snp2 8  7  87.5
	etc
	 
	Q2: From the individual crosstabs tables, can I automatically
extract those numbers in columns 2 and 3?
	 
	 
	rgds,
	jenny

______________________________________________
R-help at stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html