Hi, Kathy. I suppose I had a 50/50 chance of guessing correctly when I chose MyTrueAncestry. CRI Genetics is the other company I could never in good conscience direct anyone to for testing or evaluation.
I don't remember ending up as "anonymous" in the thread (gasp; that means my total G2G word count is even higher than I've estimated...over 2 million words 
), but here is a G2G question from 2019 about CRI Genetics and its services.
I just did a quick check, and at least CRI has been able to boost their Better Business Bureau rating up to a "B" now by improving their responses in resolving issues. That said--and the science, or semblance thereof, aside--they've had 685 BBB complaints logged in the last three years. For less-than-large player in the industry, that's an awful lot.
On the science front, the truth is that inexpensive microarray tests have limited capability to inform us about medical/clinical conditions or risks, and extremely little, if anything, about phenotypical issues like fat metabolism, or response to exercise or a specific diet.
The federal Food and Drug Administration has, in the now 21 years that we've had any form of direct-to-consumer DNA testing, approved a grand total of only four tests, all from 23andMe. And even at that, the BRCA analysis remains controversial because, while valid, it can't take into account the other various factors involved in breast cancer, so interpreted false negatives are not uncommon. The latter is also partly due to the fact that the majority of physicians--unless their specialties call for it--are still not well-informed regarding the current state of genetics research; that remains a primary focus of the American Society of Human Genetics.
The reason the microarray tests are so limited is simple: for DTC testing, the maximum number of single nucleotide polymorphisms examined numbers about 700,000, or about 0.023% of the genome. These are not protein coding genes that are examined, but merely individual DNA "letters" among the ~6.12 billion (~3.06 billion base pairs) that we have. The average size of a protein coding gene is somewhere around 2,100 base pairs; the largest one known codes for a protein called dystrophin and spans about 2.4 million base pairs.
The approved BRCA analysis from 23andMe that I mentioned consists of looking only at three specific positions on Chromosome 17: the letters at the locations 185, 5382, and 6174. But very seldom can the value of single positions like this tell us much because the more we learn, the more we've come to realize that, when it comes to genetics, it takes a village.
We're starting to look at DNA not as an intrinsic, fundamental blueprint, per se--as we did at the start of the Human Genome Project--but more as a vast collection of interacting building blocks...as being raw material rather than solely the architectural diagram.
In fact, today being DNA Day and all, before the Human Genome Project began, scientists were encouraged to place their bets on the total number of genes the project would find. The guesses ranged from more than 312,000 to just under 26,000, with an average of around 40,000. At the "completion" of the Human Genome Project in 2003 the estimate arrived at was about 35,000. Right now the current consensus, depending upon where you find it, is around 20,500. Much lower than we thought only a couple of decades ago...because DNA isn't a static blueprint, after all.
We've discovered amazing things in just the past few years, things like genome transfer via cell-to-cell travel of whole organelles being a reality; that simple activities like resistance exercise result in long-term changes through epigenetic activation and/or suppression of over 150 different genes; that our germline genomes change as we age through methylation and deamination; that determinants of transcription factor binding affinity can be identified as being structured like neural networks.
This is just a cautionary tale for the casual G2G reader. There are definite insights to be gained from the 700,000 point samples in an at-home autosomal DNA test. Things like 23andMe's BRCA analysis. However, the more any person or organization tries to tout that such point-datum information is definitive about clinical, pharmacological, or (particularly) wellness and lifestyle matters, the more I believe we should, with gusto, heave a pound of salt at the assertions and demand, "Show me the peer-reviewed research studies," before we believe a milligram of it.
The actual, peer-reviewed science is telling us to greater and greater extents that our genes work not only collectively with each other in complex ways, but that even more complex are epigenetic factors, both heritable and environmental, that result in variable histone and methylation changes, changes in RNA production, and transcriptional gene activation and suppression.
With the first, full, telomere-to-telomere sequencing of the human genome last year, there are great prospects that the hybrid sequencing methodology can start to allow better insight into heritable epigenetics and the role they may play. But we're not there yet, and we'll never get that kind of data from a microarray test: those simply can't look at the chromatin or heterochromatic regions of our genomes. Even our current commercial whole genome sequencing tests like the 30X coverage I took can't do that.
