Ambiguous flip
While at this early stage *all data is suspect*, some data is even more so. Genotypes which cannot easily be distinguished from their flipped form are very prone to confusion, including by scientists when they publish their results [PMID 18154681].
Since DNA is composed of 2 antiparallel strands, there is ambiguity over which strand to look at. dbSNP uses the assembled chromosome to establish a plus and a minus strand. Sometimes other sources rely on the orientation of an individual read, the encoded gene, or other information.
You may see the terms orientation and strand used interchangably.
If a microarray claims that you are an rs1234(A;A) for a SNP in which the other allele is G, but dbSNP claims that this is a C;T SNP, then logically we flip your results over and call you a rs1234(T;T). This is safe and reasonable.
Unfortunately if this was instead a SNP where the two alleles are A or T the same flipping logic falls down. We don't (yet) have a way to know for sure if you should be flipped or not, since both forms of your flip rs1234(A;A) and rs1234(T;T) are possible. That problem occurs for the homozygous forms of A/T and C/G SNPs, and remains hard, with several possible avenues of attack, but no clear solution.
[edit] be extra skeptical of these cases
- an (A;A) genotype call when the alleles are A and T
- a (T;T) genotype call when the alleles are A and T
- a (C;C) genotype call when the alleles are C and G
- a (G;G) genotype call when the alleles are C and G
A possible solution to this is to use the Illumina TOPBOT solution
