Hyejung Won @hyejungwon.bsky.social · 23/09/202616/ Together, our results connect genetic variation associated with cortical structure to regulatory mechanisms in developing human neural cells. 010
Hyejung Won @hyejungwon.bsky.social · 23/09/202615/ This illustrates a broader point: the functional impact of a genetic variant can depend not only on its DNA sequence, but also on the developmental signaling context of the cell. 110
Hyejung Won @hyejungwon.bsky.social · 23/09/202614/ One example was a variant near FOXO3: its allelic effect was muted at baseline but became much clearer after Wnt stimulation. 110
Hyejung Won @hyejungwon.bsky.social · 23/09/202613/ Finally, Wnt stimulation showed that regulatory effects are not always fixed. 12 variants showed significant condition-dependent allelic effects, revealing genetic effects that depend on signaling context. 110
Hyejung Won @hyejungwon.bsky.social · 23/09/202612/ In other words, the same variant may have different consequences depending on which transcription factors are available in a given region. 110
Hyejung Won @hyejungwon.bsky.social · 23/09/202611/ One clue came from transcription factors, proteins that “read” regulatory DNA. Across cortical regions, we found that regional differences in GWAS effect sizes tracked with regional expression of transcription factors predicted to interact with these variants. 110
Hyejung Won @hyejungwon.bsky.social · 23/09/202610/ We next turned to another puzzle: A genetic variant is present throughout the brain. So why can its association be much stronger in one cortical region than another? 110
Hyejung Won @hyejungwon.bsky.social · 23/09/20269/ Our results suggest that Alu sequences may have helped shape regulatory programs in the developing human brain, and that common variation within these sequences now contributes to differences in cortical structure across individuals. 110
Hyejung Won @hyejungwon.bsky.social · 23/09/20268/ Why is that interesting? Transposable elements have long been viewed as genomic remnants, but they can also provide new regulatory sequences for evolution to work with. 110
Hyejung Won @hyejungwon.bsky.social · 23/09/20267/ Then came one of the more unexpected findings: Many of these regulatory variants were embedded in Alu elements: primate-specific transposable elements that make up a surprisingly large fraction of the human genome. 110
Hyejung Won @hyejungwon.bsky.social · 23/09/20266/ Across the library, we identified 386 expression-modulating variants (emVars) across 107 loci: strong candidates for causal variants underlying cortical structure associations. 110
Hyejung Won @hyejungwon.bsky.social · 23/09/20265/ We therefore performed MPRA both at baseline and after Wnt stimulation, a key pathway controlling NPC proliferation and cortical development. 110
Hyejung Won @hyejungwon.bsky.social · 23/09/20264/ And because gene regulation does not happen in a vacuum, we asked another question: Does the effect of a genetic variant change depending on the signaling environment of the cell? 110
Hyejung Won @hyejungwon.bsky.social · 23/09/20263/ To address this, we used a massively parallel reporter assay (MPRA) to test the regulatory activity of ~9,000 cortical structure-associated variants in human neural progenitor cells (NPCs), cells that give rise to neurons of the developing cortex. 110
Hyejung Won @hyejungwon.bsky.social · 23/09/20262/ GWAS have identified hundreds of genomic regions associated with human cortical structure. But there is a major gap: how do these variants actually influence human cortical development and structure? 110
Hyejung Won @hyejungwon.bsky.social · 23/09/20261/ I’m excited to share our new work, now out in @natneuro.nature.com! This was a wonderful collaboration with my long-term collaborator @steinlab.bsky.social, led by two fantastic trainees, @nanamatoba.bsky.social and Jessica McAfee. www.nature.com/articles/s41...nature.comMassively parallel assessment of gene regulatory activity at human cortical-structure-associated variants - Nature NeuroscienceA large-scale functional screen identified noncoding variants that regulate gene activity during human brain development, providing insights into the molecular mechanisms that shape differences in hum... 2186
Hyejung Won @hyejungwon.bsky.social · 10/06/2026You are so kind @martinowk.bsky.social! Thanks so much for all your support and help! 010
Reposted by Hyejung WonKeri Martinowich @martinowk.bsky.social · 10/06/2026Congrats to Jiseok Lee and others in @hyejungwon.bsky.social lab on this set of exciting findings that begin to explain how common variants in schizophrenia mediate functional effects. Honored for our teams to have played a role and to see our new SCZ data get integrated into a study so quickly!! 3135
Reposted by Hyejung WonCedric Boeckx @cedricboeckx.bsky.social · 10/06/2026Fascinating new work linking schizophrenia to cilia By @hyejungwon.bsky.social @martinowk.bsky.social @stephaniehicks.bsky.social @kr-maynard.bsky.social “first in-vivo characterization of the consequence of common variant architecture in SCZ; ciliary dysfunction as a convergent mechanism” 🧪🧬🧠biorxiv.orgPerturbation of genes linked to common schizophrenia risk variants identifies cilia programsSchizophrenia (SCZ) is a common psychiatric disorder characterized by psychosis, emotional withdrawal, and cognitive deficits. Most SCZ risk variants reside in non-coding regions of the genome and are... 0168
Hyejung Won @hyejungwon.bsky.social · 14/04/2026Huge congrats to Sang Ho and you Keri! You two should be very proud 120
Reposted by Hyejung WonSang Ho (Sangho) Kwon @shkwon-bcmbneuro.bsky.social · 18/02/2026FIRST post here on Bluesky - we’re excited to share that our large spatial transcriptomics study of the human dlPFC in schizophrenia is now available on bioRxiv📄: www.biorxiv.org/content/10.6... (1/9) 34315
Hyejung Won @hyejungwon.bsky.social · 09/01/2026Huge congrats Keri! Well deserved, and LIBD is lucky to have you as a CSO! 110
Reposted by Hyejung WonJonathan Pritchard @jkpritch.bsky.social · 06/01/2026New preprint alert: we use sign errors as a test of how well TWAS works. Very worryingly we find that TWAS gets the sign wrong around 1/3 of the time (compared to 50% for pure guessing). You can read more about our analysis here, and what we think is going on 👇 56728
Reposted by Hyejung WonPsychiatric Genomics Consortium (PGC) @pgcgenetics.bsky.social · 03/12/2025So, who are the the Functional Genomics Working Group? 🎯Our goal is to understand how genes🧬, cells🦠and molecules💠contribute to psychiatric disorders🧠 We work with other PGC groups and different 'omics data (methylation, cytometry, single cell etc.) to do this🧪 🧵1/3 184
Hyejung Won @hyejungwon.bsky.social · 02/07/2025Thanks so much Keri for inviting me ❤️ It was a great pleasure to visit Lieber and meet so many amazing scientists! 010
Reposted by Hyejung WonKeri Martinowich @martinowk.bsky.social · 24/06/2025Phenomenal talk from @hyejungwon.bsky.social who is visiting us @lieberinstitute.bsky.social for the day. Groundbreaking work w/ important implications and roadmap for using rapidly emerging information on psychiatric genetics to prioritize variants and understand their impact on 🧠 function 2113
Reposted by Hyejung WonDoug Fowler @dougfowler.bsky.social · 15/05/2025Our "Atlas of Variant Effects 2030 Roadmap" is live: zenodo.org/records/1542... 1/nzenodo.orgAtlas of Variant Effects 2030 Roadmap: resolving human variants of uncertain significanceAt the Clinical Atlas of Variant Effects meeting (CLAVE meeting, July 2024, Pittsburgh USA), we developed recommendations for a draft atlas that can be realized by 2030, with a focus on empowering gen... 13727
Reposted by Hyejung WonSarah Marzi @sj-marzi.bsky.social · 27/05/2025How does APOE change your risk of developing Alzheimer's disease? At least partially by changing the function of immune cells in your brain 🧠🧬 New paper out in @natcomms.nature.com today: www.nature.com/articles/s41... @ukdri.ac.uk @kingsioppn.bsky.social @imperialbrains.bsky.socialnature.com 24717
Hyejung Won @hyejungwon.bsky.social · 14/05/2025Huge congrats Debby! This was a lovely story when you presented at GRC ❤️ 140
Reposted by Hyejung WonMartin Kampmann @kampmann.bsky.social · 16/04/2025Silencing of transgenes during iPSC differentiation is a frequent problem. 🧬🫢 Check out the piggyBac system below, which has resisted silencing in many protocols across many labs. 💪 Excited that our lab could contribute to this team effort led by @tuenaka.bsky.social & Marius Wernig! 02712
Reposted by Hyejung WonPGC Functional Genomics @fg-pgc.bsky.social · 11/03/2025Watch the video📽️attached to see what each of these awesome #FemsinSTEM are up to! 🧵3/3 022
Hyejung Won @hyejungwon.bsky.social · 06/03/2025Marija, I am so much happy for you! Huge congrats! 😍 110
Hyejung Won @hyejungwon.bsky.social · 28/02/2025Hello, it is because borderline disorder GWAS is not well powered: i.e. it does not have any hits (dots) above the red line in figure 2: www.nature.com/articles/tp2.... This means that there are not robust and reliable genomic regions that we can chase after for functional analysis.nature.comGenome-wide association study of borderline personality disorder reveals genetic overlap with bipolar disorder, major depression and schizophrenia - Translational PsychiatryTranslational Psychiatry - Genome-wide association study of borderline personality disorder reveals genetic overlap with bipolar disorder, major depression and schizophrenia 010
Reposted by Hyejung WonCedric Boeckx @cedricboeckx.bsky.social · 12/02/2025Very good new preprint by @nanamatoba.bsky.social @hyejungwon.bsky.social @steinlab.bsky.social, using a Massively Parallel Reporter Assay to gain insight into regulatory elements associated with inter-individual differences in cortical structure (Wnt response sensitivity plays a prominent role)🧪🧠🧬biorxiv.orgMassively parallel assessment of gene regulatory activity at human cortical structure associated variantsGenetic association studies have identified hundreds of largely non-coding loci associated with inter-individual differences in the structure of the human cortex, though the specific genetic variants ... 062
Hyejung Won @hyejungwon.bsky.social · 28/01/202515/ Last but not least, I'd like to thank fabulous first authors, Sool Lee, Jessica McAfee, and Jiseok Lee; great collaborators, Pat Sullivan, Adriana Beltran, @markgerstein.bsky.social, Alan Boyle; and a fantastic editor, Scott Behie! 020
Hyejung Won @hyejungwon.bsky.social · 28/01/202514/ There’s much more in the full manuscript, including findings on Alu elements with gene regulatory activity and CRISPR validation. Check it out here for full access: authors.elsevier.com/a/1kU9ZL7PXq...authors.elsevier.com 110
Hyejung Won @hyejungwon.bsky.social · 28/01/202513/ Together, our findings suggest pleiotropy arises from the functional properties of target genes and proteins, including prolonged gene activity during neuronal differentiation and higher protein connectivity in PPI networks. 110
Hyejung Won @hyejungwon.bsky.social · 28/01/202512/ PPI networks were a recurring theme: mDis3-bound transcription factors, mDis3 proteins, and even downstream effectors all showed higher connectivity in PPI networks. Connectivity in these networks seems critical to how pleiotropic variants exert their effects. 100
Hyejung Won @hyejungwon.bsky.social · 28/01/202511/ We also investigated protein-protein interaction (PPI) networks. mDis3 proteins were significantly more interconnected than mDis1 proteins. This implies pleiotropic proteins may amplify their effects through network connectivity, rippling through molecular pathways. 110
Hyejung Won @hyejungwon.bsky.social · 28/01/202510/ This suggests that mDis3 variants target genes with broader or longer activity across excitatory neuronal differentiation, potentially driving multiple downstream effects. These temporal differences may explain part of the functional diversity seen in pleiotropy. 110
Hyejung Won @hyejungwon.bsky.social · 28/01/20259/ We next examined the functional properties of variants and their target genes. Both mDis1 and mDis3 variants were enriched in the excitatory neuronal lineage. In this lineage, mDis1 genes were often expressed in a single cell type, while mDis3 genes showed prolonged, multi-cell-type expression. 100
Hyejung Won @hyejungwon.bsky.social · 28/01/20258/ Next, we wondered if mDis3 variants have more complex regulatory connectivity (e.g., linking to multiple genes). Again, no significant differences were found between mDis3 and mDis1 loci in the number of genes connected to emVars. 100
Hyejung Won @hyejungwon.bsky.social · 28/01/20257/ First, we asked if mDis3 loci simply have more emVars than mDis1 loci. The number of emVars didn’t differ between mDis3 and mDis1 loci. So, the extent of pleiotropy wasn’t explained by the number of variants with regulatory activity. 100
