Eric Fauman

@Eric_Fauman

Senior Scientific Director, Integrative Biology . Goal: identify causal genes for all GWAS loci. Progress so far at link below All views my own.

Vrijeme pridruživanja: prosinac 2013.

Tweetovi

Blokirali ste korisnika/cu @Eric_Fauman

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @Eric_Fauman

  1. 3. velj

    So in the region we have a very clear link of serine to PSPH and a very weak link between CHCD2 and Parkinson and no link whatsoever between the two. That means this conclusion in the paper and that implied by the title are incorrect:

    Prikaži ovu nit
    Poništi
  2. 3. velj

    Instead the authors label the locus by the proximal gene, CHCHD2. CHCHD2 has been linked to an autosomal dominant Parkinson Disease: And they report a lack of any linkage between PD variants and the serine variants

    Prikaži ovu nit
    Poništi
  3. 3. velj

    So now there's a new metabolite GWAS in a Chinese cohort. The lead SNP for serine is a little further away from PSPH, but it's in high LD with the prior signal. And even though the authors site all 3 prior studies, the paper never mentions PSPH

    Prikaži ovu nit
    Poništi
  4. 3. velj
    Prikaži ovu nit
    Poništi
  5. 3. velj

    What's wonderful about metabolite GWAS is that usually the causal gene is abundantly clear. In our 2014 Shin et al paper we described an association with serine near PSPH. The PSPH protein is the last step in the synthesis of serine. And this biology has been known for 65 years!

    Prikaži ovu nit
    Poništi
  6. 3. velj
    Prikaži ovu nit
    Poništi
  7. 1. velj

    Clearly the "effect size" of the TBC1D4 variant will depend on the diet of the population, a nice example of gene-environment interaction. Here's a related argument showing that PKU is both 100% environmental and 100% genetic What do you think?

    Prikaži ovu nit
    Poništi
  8. 1. velj

    It's possible that under a traditional diet (meat and organs of seal and other marine mammals) low in carbohydrates keeping glucose in circulation could be a good thing. Current diet is high in sugar and carbs "soft drinks with sugar, sweets, chips"

    Prikaži ovu nit
    Poništi
  9. 1. velj

    I'm also reminded of one of my favorite papers describing a Greenland specific TBC1D4 mutation which by itself accounts for 10% of all T2D cases in Greenland. The mutation causes a muscle-specific LoF of TBC1D4 driving up postprandial glucose and insulin

    Prikaži ovu nit
    Poništi
  10. 1. velj

    This interplay between genes, exposure & reference population reminded me of this recent tweet pointing out that "If everyone smoked 20 cigarettes a day, then case-control & cohort studies would lead us to conclude that lung cancer is a genetic disease"

    Prikaži ovu nit
    Poništi
  11. 1. velj
    Prikaži ovu nit
    Poništi
  12. 1. velj

    As it turns out most animals including most mammals make their own ascorbic acid. The last synthetic step is catalyzed by the enzyme encoded by GulO

    Prikaži ovu nit
    Poništi
  13. 1. velj

    I was being provocative in my reply but I find it interesting how "disease" is a social construct or at least requires a reference population. The point here is that although scurvy is listed at the epitome of an environmental disease it can also be considered a genetic disease.

    Prikaži ovu nit
    Poništi
  14. 31. sij
    Poništi
  15. 31. sij

    I'll conclude with this link to my analysis of the urinary metabolites where I think we _can_ establish the true causal genes. In that case colocalization with relevant cell-type specific eQTLs was successful in identifying ~20% of the true causal genes.

    Prikaži ovu nit
    Poništi
  16. 31. sij

    Both RDH5 and TRPM1 were selected by colocalization. Both are also closest genes. At the RDH5 locus colocalization also flagged BLOC1S1. However I can't find even a hand-waving rationale for BLOC1S1 in AMD.

    Prikaži ovu nit
    Poništi
  17. 31. sij

    A very successful approach to identifying causal genes is to look for Mendelian disease genes with highly similar phenotypes to the GWAS trait. Here I find 2: RDH5 causes fundus albipunctatus TRPM1 causes night blindness

    Prikaži ovu nit
    Poništi
  18. 31. sij

    However the colocalization analysis identified HTRA1 as the causal gene at this locus. I found one paper providing a rationale for linking HTRA1 to AMD but as you can see in the diagram it provides equal support for ARMS2 .

    Prikaži ovu nit
    Poništi
  19. 31. sij

    The lead ARMS2 SNP is in high LD with a missense variant (A69S) and an indel and it apparently plays a role in the complement cascade.

    Prikaži ovu nit
    Poništi
  20. 31. sij

    One locus that caught my eye was ARMS2-HTRA1rs3750846 with a neglogp of 734 for AMD. ARMS2 actually stands for "Age-related maculopathy susceptibility protein 2". This just means historically (2005!) this locus was associated with AMD

    Prikaži ovu nit
    Poništi

Čini se da učitavanje traje već neko vrijeme.

Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.

    Možda bi vam se svidjelo i ovo:

    ·