ksuhre.bsky.social
Passionate about #GWAS #Metabolomics #Proteomics #Glycomics #Epigenomics, Professor @WeillCornell
Medicine - Qatar, Blog on http://metabolomix.com, http://suhre.fr
67 posts
1,372 followers
2,829 following
Prolific Poster
Conversation Starter
comment in response to
post
The key correlation plots are here:
analyticalsciencejournals.onlinelibrary.wiley.com/action/downl...
comment in response to
post
I agree that CHRNA2 is the best candidate to explain smoking behavior, but would bet that there is some co-evolution between the regulation of the two smoking related genes, assuming that apes did not smoke.
comment in response to
post
Would be interesting to look at the regional association plots for the eQTLs and sQTLs for the two genes, but can’t do that right now on my iPhone
comment in response to
post
Although epoxide hydrolase 2 also can be linked to smoking, so maybe two causal genes under a shared promoter variant?
comment in response to
post
My guess is cholinergic receptor nicotinic alpha 2 subunit, given the lead association of this variant with smoking … info from genetics.opentargets.org/Variant/8_27... helped.
comment in response to
post
Please do … fell behind during the last year!
comment in response to
post
The weakest correlations were due to detectability variations between platforms in targets like IL-5, IL-13, IL-2RB, and IL-20 from cases. These targets had above 95% detectability in Alamar and around 10% detectability in Olink, which could arise from the different antibody types used per assay.
comment in response to
post
Thank you - will do ... didn't update it in a while, so it's time!
comment in response to
post
The key correlation plots are here:
analyticalsciencejournals.onlinelibrary.wiley.com/action/downl...
comment in response to
post
Yes - Spotify style would also be OK, a business model that rewards quality rather than quantity.
comment in response to
post
Humans as the best preclinical models ?
comment in response to
post
Are you into generating captchas now 😉
comment in response to
post
So it's approx 300 million Eur just for Olink kits 😏 plus sequencing and labor... Nice.
comment in response to
post
This level of commitment also gives some insight into the value that the 50k “pilot” must have generated for the UKB-PPP participants
comment in response to
post
Or DM / email me and I can put you in contact with my collaborators at SEER who developed the python scripts to implement the method we used in the paper
comment in response to
post
I could share the FASTA, it contains cohort specific variants down to MAF = 10%
comment in response to
post
For sone time I maintained a sourdough starter using a protocol to make a sourdough bread per day (in a country where this was rare to come by), but getting consistent results was tricky, putting all ingredients into the machine in the evening and programming it for baking the next morning.
comment in response to
post
Here an idea - use Zoom images to predict flu waves in real time, a bit like they did using Google searches
comment in response to
post
Although Eddie Park and Yi Xing argue that their results should not apply standard splicing QTLs (if I understand them correctly), we definitely see very similar effects with sQTLs as well. Here's one example from our Text S2: