Scientists have uncovered hidden genetic drivers of conditions like IBD by stimulating immune cells. 🔬
The findings have resulted in a dataset called MacroMap, which offers insight into why some people are vulnerable to certain conditions.
sanger.ac.uk/news_item/ac...
...clinical collaborators (Tim Raine and @gastrogrj.bsky.social ) and donors and their families! 17/
Huge thanks to the many members of the @carlanderson.bsky.social team who drove this work (past and present), our funders ( @opentargets.org @crohnscolitisfdn.bsky.social @wellcometrust.bsky.social), @sangerinstitute.bsky.social, ...
If you think this is the coolest thing ever (like I do) then good news 🥳 we are hiring! So do get in touch with myself or @anderson_carl if you’re interested in using genetics and population scale, longitudinal scRNAseq to understand common complex diseases! andersonlab.info 16/
🍊The really juicy biology, including deep dives into specific hits and what this could mean for drug development, is summarised by @Tobionformatics here 💊. There’s some very interesting and surprising findings, and even more in the paper - so do check it out! 15/
bsky.app/profile/tobi...
These effects can therefore only be captured by using scRNAseq. However, even at this scale, there are many cell-types for which we are relatively underpowered. Continued high resolution eQTL mapping in ever larger datasets will likely continue to help understand GWAS hits. 14/
However, because we find this substantial enrichment at higher resolutions, we believe effector gene dysregulation is largely contextually restricted. Such effects may therefore hide from selective pressures, and persist at the common frequencies often found by GWAS. 13/
🤔 We think this makes a lot of sense evolutionarily: If disease effector gene dysregulation is widespread (i.e more detectable at lower resolutions), it may yield a greater phenotypic effect, and be influenced by greater selective pressures. 12/
So are those effects found at each resolution equally likely to drive disease❓ NO - Those eGenes found at the cell-type level were SUBSTANTIALLY enriched for disease effector genes - a whopping ~3.5-fold (2.68/0.75) more so than the ‘All Cells’ level. 11/
To see which of these underpin susceptibility, we colocalised these with IBD GWAS. Remarkably, we nominate effector genes at an enormous 74 (❗) loci where one has not previously been nominated in @OpenTargets. This therefore SUBSTANTIALLY improves on previous efforts. 10/
Something really cool! 🌟 While most genes have an eQTL at the ‘All Cells’ level (‘eGenes’), many eQTLs were found only at higher resolutions. So while rarely finding new eGenes, we find many new regulators. These are further from the TSS and more likely found in enhancers. 9/
🗺️ We then mapped eQTLs at several resolutions;
1) ‘All Cells’ (like bulk)
2) Major populations (coarse resolution)
3) Cell-types (high resolution)
Doing this within or across anatomical sites, we find >84k eQTLs (❗) in 251 different annotations. 8/
To answer this, we generated ‘IBDverse’ 💫 - The world’s LARGEST collection of scRNAseq data from the sites most relevant for IBD!
🤯 Across this gigantic set of 2.2M cells from 732 samples, we identified 9 major populations that comprised 86 cell-types. 7/
Using inflammatory bowel disease (IBD) as an example, we wanted to test:
🔍 What kind of eQTLs can we identify if we preserve the cellular resolution of expression by using single-cell RNAseq (sc-eQTLs)?
❓Do these better nominate disease effector genes. 6/
😥 Unfortunately these have had little success. Most work, however, relies on bulk RNAseq, which requires homogenising a sample much like you do to fruit in a smoothie. But what if disease-causing gene dysregulation is missed by doing this? 5/
⏩ This led to widespread efforts to identify regulatory variants, a.k.a ‘expression quantitative trait loci’ (eQTLs), that colocalise with GWAS signals, in the hope of pinpointing the effector genes, tissues and cell-types. 4/
However, ~90% of GWAS hits lie outside genes themselves. While this makes it difficult to nominate the ‘effector gene’, it indicates these likely act by modifying their expression level - this can be quantified by measuring the abundance of the associated RNA. 3/
🕵️Genome wide association studies (GWAS) seek to identify genetic variants that drive complex traits and diseases. Because we can be fairly confident these variants are causal of the phenotype, drugs targeting associated genes are much more likely to be successful. 2/
Delighted to share the first preprint of my postdoc in the @carlanderson.bsky.social lab www.medrxiv.org/content/10.1...! 🚨 A super exciting study I co-led with the very talented @tobioinformatics.bsky.social . Stay tuned to see what we learned about genetic susceptibility to complex disease. 🧬🧵 1/
UMAP of the 9 populations and 86 cell types identified after quality control and clustering
Preprint out today!
A team led by @tobioinformatics.bsky.social and Bradley Harris in @carlanderson.bsky.social ‘s lab has created the largest single-cell atlas of IBD tissues to date
www.medrxiv.org/content/10.1...
I never caught one this nice though!