SNP "Flip"/"Invert" and Sequencing

[cpurick]

I'm working with raw SNP data from Embark (think 23&Me for dogs).

I looked up a genetic mutation in a university study, and some (not all) of the SNPs in the report are inverse (A<-->T and C<-->G) to the raw data from Embark. Is there a standard for this? If the study found T at a specific SNP, how do I know whether an A detected by Embark is a different variant, or simply the same base pair viewed from the other strand? I would think they could write a probe to get the same SNP from whichever strand they want, but then they would have to know which value correlates to each variant for their particular probe, wouldn't they? Would it be common for unrelated labs/tests to study the same base pair from opposite strands?

Also, I get heterozygous SNPs like AG then AG. But I presume there is no telling whether these peptides are actually sequenced AA and GG on the strands or AG GA, correct? Taking that a step further, it would appear to actually be impossible to actually sequence from SNPs simply because they do not distinguish strands. Is that correct?

Edited June 21, 2019 by cpurick
typo