I've got some human GWAS data and looking at rs7329280 (chr13), which seems to be highly associated with the sex. This doesn't make much sense to me, it's an autosomal SNPs and on UCSC browser there's not much annotated in that region (besides the gene itself, TPTE2).
Is it likely that this SNP is either (1) in long range LD with the sex chromosomes or (2) that the probe sequence has some non-unique binding so it's capturing some sex chromosome region?
I've used blastn to check the sequence (TTCCTTGGACCTCCTAGTCAGGTCACGTAATGACGTTTCTCTGGGAATAA) against human references, the top hits seem to indicate the second option, is that right:
|Select seq ref|NC_018924.2|||99.6||470||100%||3e-19||100%||NC_018924.2|
|Select seq ref|AC_000145.1|||99.6||470||100%||3e-19||100%||AC_000145.1|
|Select seq ref|NC_000013.11|||99.6||531||100%||3e-19||100%||NC_000013.11|
|Select seq ref|NC_000024.10|||83.8||253||100%||2e-14||96%||NC_000024.10|
How do people filter these SNPs out, are there lists of known "culprits" somewhere?