Genetic testing - nutrigenomics? Nutrahacker, Genetic genie, etc.

[kathy775]

Just wanted to share this in case this is of interest or of any help to anyone here. (My family member had PD, I am not currently diagnosed with it; I'd been gathering info on his behalf here in the past.)Recently I watched a podcast with Bryan Johnson, a billionaire who is aiming to reverse his biological age ... a biohacker. Controversial figure, but he's very systematic and dedicated in his approach. Basically he says he lets the data speak for itself in determining his diet, supplements, exercise, to try to stave off effects of aging.

I myself have tried many supplements ... very trial and error ... but in the spirit of "let the data speak," I finally decided to use data from a 23andme genetic test I'd taken years ago and just got a nutrigenomic test done (Nutrahacker; there are others).

I found I have MTHFR variants and slow COMT, among other things, and neurotransmitter levels, ability for liver to detox, glucose metabolism, etc., etc. Certain supplements were recommended, some were contraindicated to help influence genetic expression positively. Found I was taking some things I should actively avoid. It was eye opening. One can say a supplement or food is healthy, but it may really depend on genotype.

Anyway, such testing might be a helpful point of reference. One can upload raw data from 23andme or Ancestry, etc. Genetic Genie doesn't give suppement recommendations but it offers metabolic and detox profile reports (free for now).

All the best.

[genesurf]

Also-- If you're eligible, the PD GENEration will analyze your genome for Parkinson's variants for free. The under-$300 consumer genetic tests do not sequence one of the important PD genes (GBA) correctly, but the PD GENEration project does.

For the GBA gene, if you have a whole genome sequence (WGS) like Dante or Nebula, you can try running your data through Gauchian. Gauchian is a GitHub program that uses statistics to try to correctly identify GBA mutations.

[kathy775 in reply to genesurf]

Excellent, thanks so much for these details in this very informative reply. I'm in Canada and I gather the PD GENEration is only in the U.S. However, I was starting to consider any possible covered genetic testing/counselling for PD where I am in Canada and found two local hospitals that do this, if I'm eligible, but I need to be referred by my doctor. And I was considering WGS but prefer free testing if available.

I may possibly share this info on another forum, it's very helpful. Many thanks!

Regards,

Kathy

[genesurf in reply to kathy775]

I think you're better off with the official targeted tests if they are covered. 30x WGS can't read the GBA gene correctly-- during processing , many of the GBA gene fragments are mapped to the GBAP1 gene and vice versa. The genes are nearly identical, but they have different functions.

Researchers have addressed the WGS problem in this paper: nature.com/articles/s42003-..., and they developed a python program called Gauchian: github.com/Illumina/Gauchian. If you are tech savvy you can run your BAM or CRAM file through Gauchian, and it will analyze your GBA and GBAP1 genes, and try to determine the "true" values. But a targeted test is going to be more accurate.

[kathy775 in reply to genesurf]

Thank you for the distinctions made here in the accuracy of testing, Genesurf. I am duly impressed with your understanding of this.

I'm afraid I myself am not tech savvy at all, so I will see if I can go down the route of targeted testing. Thanks again!!

[nop9 in reply to genesurf]

thanks for the info. i checked my bam file from sequencing.com. had to convert it (1->chr1) and add an index with samtools. everything negative, just: CN(GBA+GBAP1):4 gensurf, do you know what it means?

[genesurf in reply to nop9]

The CN(GBA+GBAP1) field shows total copy number of the two combined, as quoted below. I had "4" for almost all my files. More details on Gauchian are available here: ncbi.nlm.nih.gov/pmc/articl... "... A deviation of this copy number (CN) from the diploid expectation indicates the presence of a CNV, e.g. one copy indicates a CNL, and three or more copies indicate a CNG. Thus, this number plus two gives the total copies of GBA and GBAP1 combined, i.e. CN (GBA + GBAP1). Included in this CN calculation, in addition to GBA and GBAP1 genes, are gene hybrids where part of GBA and GBAP1 are fused..."

For some of my files I had erroneous results from Gauchian when I accidentally tagged the wrong genome build. I knew something was wrong because the .json showed SNPs with "0,0" instead of expected values like "0,28". So be sure to check that everything looks normal there.

Finally, this was just recently published, it says that Gauchian is good but not perfect so it can't be used diagnostically: medrxiv.org/content/10.1101... "While Gauchian genotyped most patients correctly, it missed some rare or de novo mutations due to its limited internal database and over-reliance on intergenic structural variants. This resulted in misreported homozygosity, incomplete genotypes, and undetected recombination events, limiting Gauchian’s utility in variant screening and precluding its use in diagnostics."

[nop9 in reply to genesurf]

thank you very much for this detailed description!