How To Perform Gwas Analysis With Special (Single) Marker?

Question

[xuanbonn]

Hi all,

I was wondering if there is a software available for special gwas analysis. Special means: instead of biallelic markers, this software will use single marker. For example, in PLINK, the input is famID, OffsprintID fatID MoID Sex Affection alleles

1        1               0       0      1     2           **A T**
1        2               0       0      2     1           **A A**
1        3               1       2      1     2           **A T**

I want something like following to do association analysis: famID, OffsprintID fatID MoID Sex Affection allele

1        1               0       0      1     2           **A** 
1        2               0       0      2     1           **A** 
1        3               1       2      1     2           **A**

Someone said it is OK to add a pseudo marker next to single marker, this pseudo one should be the same as real one. At the end ,plink will be able to do the analysis. However I think this will change the allele frequencies,right?

Any idea?

[matted]

What exactly are you asking? Are you asking about variants on the X chromosome for males, where there is only one allele? Or are you working with a haploid species?

Read the plink documentation for X chromosome markers: http://pngu.mgh.harvard.edu/~purcell/plink/faq.shtml#faq9.

We (and you?) need to know more about what you're trying to do in order to answer this. For example, for simple linear association tests, it won't make a difference what you do. If you are filtering by HWE or something, doubling the alleles might give you trouble.

[xuanbonn]

Thanks for reply. I am asking whether doubling alleles will influence the result of standard case/control study that compares allele frequencies. Because as I understood, standard case/control study will consider such table:

 Group                                        Genotype Count (Percent)                                              
                                   E4/E4             E4/+                +/+             Total

Alzheimer's Disease patients      12 (13%)            46 (50%)          33 (36%)           91

Unaffected controls                  2 (3%)            14 (19%)          55 (79%)           71

For each individual, phased genowide genotype data have been separated into two haploids. For the single SNP association study, I only want to take one haploid from one individual to analysis. Therefore, I am thinking to double the alleles. Once alleles changed, the above table will be changed which may lead to wrong result, or?