Hapmap wrongly called APOE SNP rs429358???

[ymc]

For the CHB population, hapmap r28 shows TT for all the genotyped 45 individuals.

But I checked the 1000genomes 20101123 release, 41 out of the 45 are genotyped and 8 of them are CT and the other 37 are TT.

It is known that rs4420638 is highly linked with rs429358. The calls for rs4420638 are consistent for both hapmap and 1000genomes.

If you go to pubmed, you can find that the APOE genotype frequency in Chinese is about 7%. This plus the rs4420638 calls make me think that the calls made by 1000genomes 20101123 release are more likely to be correct.

I also noticed the Illumina Omni1 and Omni2.5 chips called these 45 individuals all TT's (according to ftp.illumina.com). But their calls for rs4420638 is again agree with hampap/1000genomes. Since hapmap used illumina's chips for genotyping, I suspect that the rs429358 probes on the illumina chips are defective. This in turn caused the hapmap wrong calls.

I think this is a pretty serious matter given the importance of rs429358.

Can someone linked to hapmap/1000genomes or illumina rep clarify what is going on here?

Thanks a lot!