Don’t Take Any Wooden Nickels
DNA tests that tell customers what they want to hear are a disservice.
After pioneering a distinctive ancestry method centered on our population database atDNA, starting more than 20 years ago, DNA Consultants has had surprisingly few competitors. One newly minted operation that has recently entered the fray, however, deserves mentioning. We want to use this space to warn people about accepting questionable results and comparing our reports with the offerings of certain imitators. We don’t like seeing anyone misled in the DNA testing marketplace.
Under the rubric of Caveat Emptor, let us list some of the disreputable practices in forensic ancestry testing that have come to our attention just this year.
- Manipulating data. A company to avoid is any outfit that plays fast and free with the numbers. In science, the data are sacrosanct. DNA Consultants preserves the published STR allele frequencies reported from forensic population studies exactly as they appeared in print (subject to any corrections or updates). It does not pad the numbers or trim the evidence. Several Nature magazine articles this year have warned about the perils of medical studies written by chatbots, or AI. The manipulation of facts and givens can be hidden and subtle, but it is always wrong, even unethical.
- Worse than manipulating is falsifying data. One company we examined merged or “pooled” different studies and even “simulated” populations to get desired results. One or two subjects added to another strictly formed population constitutes contamination, not expansion. We never make up data. As we have said repeatedly, the beauty of DNA Consultants’ method is that we match you to scientifically reported contemporaneous, real populations. Customers want to be descended from an ancestor, not an algorithm.
- Salting the field. Just because ancestry results are pleasing doesn’t mean they are right. With most formulas in population matching you can get the probability of a result down to the smallest decimal. DNA Consultants usually has a cutoff point (one considered “significant”) at the top 10 percentile of possible populations. For 500 possible populations, the benchmark is fifty. Other companies we have looked at seem to give customers hundreds of “top” matches. A customer can pick and choose what to value. This stratagem of deceit is similar to gemstone mines at vacation spots such as in the Smokey Mountains. The operators make sure to “salt” some obvious shiny prizes in with the dross.
- GIGO. The accuracy and reliability of our results cannot be compared with those of unscrupulous imitators. If you put in incommensurate data, your get very mixed, irrational results. Garbage in, garbage out. For instance, the insertion of “minimal allele frequencies” for where there is no value in the first place in the data set tends to deflate the saliency of statistics and produce false positives. The significance of low or absent frequencies is lost. A good program should tell you not only where you likely have ancestry but where you probably do not have ancestry. Our newest competitor blurs the borders between the two. Takers of a “permissive” test may go around thinking they are Melungeon or Cherokee or Pamunkey or some other exotic ethnicity that has struck their imagination but is not plausible.
- Confusion of methods. Our competitor seeks to upsell customers to purchase a costly full genome test, which can be uploaded to Gedmatch and further manipulated to get desired results. In our experience, a proper forensic analysis takes precedence and is more comprehensive than all other tests, though it is not the end-all-and-be-all of testing. The non-conformity of genomic builds, haplotype sequencing and percentage estimates poses problems. All of these can yield variable and shifting conclusions, especially between different companies. DNA Consultants does not change its mind about what is valid. Our results are conclusive. When you have taken the DNA Fingerprint Plus there is nothing to upgrade to. Of course, new populations are occasionally added. But results are consistently sound, completely replicable and repeatable, as good science always should be. We do not believe in “the upgrade game.”
So those are reasons to raise an eyebrow at unscientific and unprofessional companies—and reasons all the more to trust the original developer, DNA Consultants. We have been on the cutting edge of forensic ancestry since 2003. Don’t accept any counterfeits.
We welcome your comments!