I am involved in a study where, as in most of life, men demonstrate themselves to be recalcitrant. So while we have many probands and most of their mothers we only have about 50% of the trios being complete. I have been running tdt and trio.types. It appears as if it is ignoring the duos. Sometimes a duo can be informative. For instance Father ..missing Mother 1/2 Proband 1/1 This duo shows that for allele 2, this was clearly a case where 2 was untransmitted. Yet I do not think this family counts toward the output that is generated. Am I correct? How do I use R to do TDT analysis where duo's are used if they are informative? If you want further details... I made a subset that I called MessWith and it is made up of the first 24 probands and their parents. 2 probands had neither a mother nor a father. Of the remaining, probands, 16 only had one parent.> summary(Genotype.914186)Number of samples typed: 52 (72.2%) Allele Frequency: (2 alleles) Count Proportion 1 73 0.7 2 31 0.3 NA 40 NA Genotype Frequency: Count Proportion 1/1 26 0.5 1/2 21 0.4 2/2 5 0.1 NA 20 NA Heterozygosity (Hu) = 0.4225168 Poly. Inf. Content = 0.3309022 tdt(Genotype.914186, MessWith, famid, pid, fatid, motid, sex, affected ) Transmission/disequilibrium test Data: Genotype.914186 Untransmitted allele frequencies, informative transmissions and exact P-values Allele Frequency Transmitted Untransmitted P-value 2 0.3333 3 2 1.000 trio.types(Genotype.914186, MessWith, famid, pid, fatid, motid, sex, affected ) Farrel Buchinsky, MD Pediatric Otolaryngologist Allegheny General Hospital Pittsburgh, PA ********************************************************************** This email and any files transmitted with it are confidentia...{{dropped}}
Using only informative incomplete trios and dropping others turns out to bias the TDT (Curtis & Sham 95). There are a variety of modifications to handle case-parent trios, starting (I think) with Weinberg's 1999 paper that uses the EM algorithm and a likelihood-ratio test. I don't know of any of these methods that are implemented in R. Perhaps they are in the package that you are using but forgot to name. Steve Buyske At 8:26 PM -0400 4/26/06, Farrel Buchinsky wrote:>I am involved in a study where, as in most of life, men demonstrate >themselves to be recalcitrant. So while we have many probands and most of >their mothers we only have about 50% of the trios being complete. > >I have been running tdt and trio.types. It appears as if it is ignoring the >duos. Sometimes a duo can be informative. For instance >Father ..missing >Mother 1/2 >Proband 1/1 >This duo shows that for allele 2, this was clearly a case where 2 was >untransmitted. >Yet I do not think this family counts toward the output that is generated. >Am I correct? How do I use R to do TDT analysis where duo's are used if they >are informative? > >If you want further details... > >I made a subset that I called MessWith and it is made up of the first 24 >probands and their parents. 2 probands had neither a mother nor a father. Of >the remaining, probands, 16 only had one parent. >> summary(Genotype.914186) > >Number of samples typed: 52 (72.2%) > >Allele Frequency: (2 alleles) > Count Proportion >1 73 0.7 >2 31 0.3 >NA 40 NA > > >Genotype Frequency: > Count Proportion >1/1 26 0.5 >1/2 21 0.4 >2/2 5 0.1 >NA 20 NA > >Heterozygosity (Hu) = 0.4225168 >Poly. Inf. Content = 0.3309022 > >tdt(Genotype.914186, MessWith, famid, pid, fatid, motid, sex, affected ) > Transmission/disequilibrium test >Data: Genotype.914186 > >Untransmitted allele frequencies, informative transmissions >and exact P-values > > Allele Frequency Transmitted Untransmitted P-value > 2 0.3333 3 2 1.000 > >trio.types(Genotype.914186, MessWith, famid, pid, fatid, motid, sex, >affected ) > >Farrel Buchinsky, MD >Pediatric Otolaryngologist >Allegheny General Hospital >Pittsburgh, PA > > >********************************************************************** >This email and any files transmitted with it are confidentia...{{dropped}} > >______________________________________________ >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
Steve Buysek Wrote
Using only informative incomplete trios and dropping others turns out to
bias the TDT (Curtis & Sham 95). There are a variety of modifications to
handle case-parent trios, starting (I think) with Weinberg's 1999 paper that
uses the EM algorithm and a likelihood-ratio test. I don't know of any of
these methods that are implemented in R. Perhaps they are in the package
that you are using but forgot to name.
Steve Buyske
My response:
I read through some of literature on the topic.
The latest publication appears to be
Guo CY, DeStefano AL, Lunetta KL, Dupuis J, Cupples LA. Expectation
maximization algorithm based haplotype relative risk (EM-HRR): test of
linkage disequilibrium using incomplete case-parents trios. Hum Hered
2005;59(3):125-35.
In it the authors make statements that lead me to believe that it would meet
my needs better than the EM likelihood-ratio test. [I am not equipped to
either agree or disagree with their assertion] The package that I was using
is dgc.genetics and as far as I know is the only tdt implementation in R.
Farrel Buchinsky
**********************************************************************
This email and any files transmitted with it are confidentia...{{dropped}}