iiinova — where your digital self keeps evolving

Reposted by Sebastiaan (Seb) Vanuytven

Dan Engelman @danengelman.bsky.social · 14/10/2026

I’m delighted to share our new paper, published today in @nature.com : “Sexual dimorphism in epidermal stem cell dynamics.” doi.org/10.1038/s415...

133 3

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 12/10/2026

🧪 LabCompass flips the script: generative AI proposes culture recipes to steer blood cell fate 🧬 mtDIVE shows which mtDNA mutations you carry — and in which metabolic context — matters more than how many 🧠 Cancer cells may "learn" drug resistance via AP-1-driven exploration + epigenetic memort

000 0

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 12/10/2026

🕸️ SpiderNet decomposes spatial data into directed "meta-interaction" modules that predict perturbation responses, brain aging & immunotherapy outcome 🦠 Env-less retrotransposons invade the germline with a virus-like fusogen

110 1

Reposted by Sebastiaan (Seb) Vanuytven

Itai Yanai @itaiyanai.bsky.social · 12/10/2026

Postdoc opening: My lab at NYU Langone Health is looking for a new postdoctoral fellow with wet-lab experience for an exciting project.

22622 26

Reposted by Sebastiaan (Seb) Vanuytven

bioRxiv Bioinfo @biorxiv-bioinfo.bsky.social · 22/09/2026

DeepSCENIC: transfer learning from sequence-to-function models enables causal gene regulatory network inference www.biorxiv.org/content/10.64898/20…

0116 11

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 20/09/2026

🎯 SIMPlex captures matched spatial & nuclear transcriptomics from a single tissue section 📍 Spatiotemporal lineage tracing reveals subclonal dynamics driving EMT & metastasis 🧠 Xenocortical mice enable in vivo study of human neural development & disease

000 0

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 20/09/2026

🗺️ SCIGMA integrates spatial multiomics across modalities with uncertainty estimates & scales to >1M locations ⚡ BatchRefiner refines single-cell embeddings in seconds, improving batch integration ✂️ Scarless conditional sgRNAs restore native guide architecture for precise CRISPR editing

110 1

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 13/09/2026

💊 Yeast-based production of etoposide/teniposide offers sustainable drug supply 🔍 Spatial multi-omics reveals dormant colorectal cancer micrometastases 🌱 Transitional metastasis stage defined by epithelial diversification & neutrophil remodeling

000 0

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 13/09/2026

🧬 miR-Space enables spatial microRNA profiling at single-cell resolution 🎨 SQUINT learns discrete cell & niche codes for spatial transcriptomics 🔄 Pik3ca H1047R drives early epithelial plasticity & CAF emergence in breast cancer ⚠️ TGFβ blockade can trigger hyperprogression in SCLC

110 1

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 07/09/2026

🧬 Deep learning decodes human regulatory DNA from single mutations 🔍 ClustoCell: 92% accuracy in reference-free single-cell annotation 💊 Semaglutide extends lifespan & slows ageing in mice 🧪 Organoids reveal distinct fibroblast roles in pancreatic cancer vs. pancreatitis

010 1

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 07/09/2026

Weekly reads 31/08/26: How AI, organoids, and GLP-1 agonists are unlocking new frontiers in biology and medicine behindthebioinformatics.substack.com/p/weekly-rea... #Genomics #SingleCell #Ageing #CancerResearch #ComputationalBiologyy

110 1

Reposted by Sebastiaan (Seb) Vanuytven

Inigo Martincorena @imartincorena.bsky.social · 03/09/2026

Big paper alert! This project has been years in the making. A comprehensive description of somatic mutation rates and signatures across 53 tissues or cell types using NanoSeq. A large collective effort led by Mimy Pham, Mike Stratton and @r-rahbari.bsky.social. www.biorxiv.org/content/10.6...

biorxiv.org

A comprehensive atlas of somatic mutation rates and mutational signatures in normal human cells

Over the course of a lifetime, somatic mutations accrue in normal human cells, causing variation in cell phenotype and engendering somatic evolution with outcomes ranging from the adaptive immune syst...

05027 50

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 23/08/2026

🧵 Mitotrek traces innate immune lineages via mtDNA 🔬 SpaMTP integrates metabolomics & transcriptomics 🔄 ZFP36L2: Stress-plasticity switch in regeneration & cancer

000 0

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 23/08/2026

🗺️ Spatial p-EMT ecosystems in HNSCC (TGFβ edge vs. OSM core) 🧬 CRISPR reveals T cell barriers: P2RY8-Gα13 & GNAS 🤖 TERRA: Zero-shot spatial transcriptomics on 112M cells

100 1

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 16/08/2026

🤖 JEPA-DNA shifts genomic models from tokens to latent semantics, boosting functional predictions. 🔄 MIRACLE enables continual, scalable integration of single-cell multimodal data 🌍 Chromatin diversity across eukaryotes: repressive states vary wildly, driven by evolutionary arms races with TEs.

011 1

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 16/08/2026

🔥 Perineural invasion in PDAC creates immune-suppressive niches (myCAFs + lipid-associated macrophages), mapped via AI tool Nicheverse. 🛡️ Immunosuppression in skin cancer disrupts macrophage-T cell crosstalk, reducing clonal diversity & worsening outcomes.

111 1

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 16/08/2026

🧬 Pan-cancer enhancer atlas uncovers 5 subgroups—EpiC4 links poor survival to FN1-CAV1-SRC-PI3K-AKT signaling, hinting at new epigenetic targets. 📍 Spatial Hi-C-RNA co-maps chromatin + transcriptome in tissues, revealing hidden tumor subdomains & developmental dynamics.

101 1

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 09/08/2026

🌐 PULSAR: A multi-scale foundation model linking molecular, cellular, and clinical biology for disease classification. 🧬 AIR-seq: RNA dynamics without chemical conversion, using NHC to mark new RNA 🎨 scVision: Turning transcriptomes into images for interpretable, zero-shot single-cell analysis.

010 1

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 09/08/2026

💰 PIP-seq: Microfluidic-free, cost-effective single-nucleus ATAC/multiome profiling, 10x cheaper than 10x Genomics. 🤖 Cell-JEPA: Latent-space prediction for robust single-cell representations, outperforming reconstruction-only models.

110 1

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 09/08/2026

🧠 TMEformer: A spatial foundation model for tumor-microenvironment interactions, predicting cellular responses in silico across cancer types. 🍔 Dietary twist: High-fat diets worsen cachexia in Lkb1-mutant lung cancer via PGE₂ signaling to sensory neurons.

110 1

Reposted by Sebastiaan (Seb) Vanuytven

bioRxiv Genomics @biorxiv-genomic.bsky.social · 05/08/2026

Multi-scale modeling of human tissues from spatial transcriptomics with TERRA www.biorxiv.org/content/10.64898/20…

032 3

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 19/07/2026

🦠 Gastric bacteria hack host epigenomes: Streptococcus anginosus delivers a chromatin remodeler to stabilize PD-L1, driving immune evasion in gastric cancer 💊 Antibiotics fuel bacterial cooperation: Sub-lethal doses induce protein transfer, boosting survival of persister cells

010 1

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 19/07/2026

🧠 Sex-dimorphic neural memory: Male mice retain long-term pancreatic plasticity via CGRP+ neurons 🧬 Inhibiting TRIM28 forces leukemia cells into terminal neutrophil differentiation 🧹 RT&T-AMP-MERFISH: Whole-transcriptome, isoform-resolved spatial imaging in intact tissue

110 1

Reposted by Sebastiaan (Seb) Vanuytven

Luca Pinello @lucapinello.bsky.social · 17/07/2026

1/ I'm excited to share that we're launching NECB 2026, the inaugural New England Computational Biology Symposium. Oct 1-2 at Microsoft Research New England, Cambridge. Two days of keynotes,talks, and posters to bring our community together across institutions.Space is limited. newenglandcompbio.org

11810 18

Reposted by Sebastiaan (Seb) Vanuytven

Ming Tommy Tang @tommytang.bsky.social · 13/07/2026

Single cell multiomics unravel the transcription networks controlling the different EMT tumor states www.nature.com/articles/s4...

031 3

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 13/07/2026

Personally, this project meant a lot to me, getting to help guide and support the team through this work was something I'd always hoped to do in science, imposter syndrome and all. #LeadByExample #LeadByKindness (19/20)

120 2

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 13/07/2026

This was a huge team effort combining scRNA-seq, scATAC-seq, CRISPR screening, and in vivo functional work. Massive thanks to all authors and to @cedric-blanpain.bsky.social for the trust and guidance throughout this project. (18/20)

100 1

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 13/07/2026

Bottom line: Klf5/Pitx1 are gatekeeping the epithelial-to-hybrid transition and enable metastasis, while Nfatc1/Creb3l1 are stabilizing later mesenchymal states. These mechanisms seem to be conserved across mouse and human cancers. (17/20)

100 1

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 13/07/2026

We also compared our mouse EMT signatures to human pan-cancer EMT programs (Gavish et al., Nature 2023). The mouse late EMT state closely matched the human "complete EMT" program, and PITX1/KLF5 vs. NFATC1/CREB3L1 activity patterns were conserved too. (16/20)

100 1

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 13/07/2026

Is this specific to skin cancer? We re-analyzed a public mouse pancreatic adenocarcinoma dataset, and found the same logic: Klf5 marks epithelial/early states, Nfatc1 and Creb3l1 mark late EMT. Same regulatory grammar, different tumor type. (15/20)

100 1

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 13/07/2026

Bulk RNA-seq after each knockout confirmed the predicted target genes: Pitx1/Klf5 loss reduced epithelial genes (Krt14, Itga6...) and boosted mesenchymal ones (collagens, Pdgfra/b), while Nfatc1/Creb3l1 loss did the reverse. (14/20)

100 1

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 13/07/2026

Unlike Pitx1/Klf5, though, Nfatc1 and Creb3l1 knockout didn't reduce metastatic burden. So late-EMT regulators seem to stabilize the mesenchymal state rather than drive dissemination, a nice functional split between the two TF modules. (13/20)

110 1

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 13/07/2026

On the other side, Nfatc1 and Creb3l1 were predicted (and confirmed by immunostaining) to be active specifically in hybrid/late EMT cells. Knocking them out blocked progression to late EMT, cells got stuck in earlier, more epithelial states. (12/20)

100 1

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 13/07/2026

Even more striking: Pitx1 and Klf5 knockout tumors formed far fewer lung metastases. So these "early EMT" TFs aren't just markers of an epithelial state, they're functionally required for efficient metastatic colonization. (11/20)

100 1

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 13/07/2026

Using CRISPR/Cas9, we knocked out Pitx1 and Klf5 in tumor cells. Both are normally expressed in epithelial + early hybrid EMT cells. Their loss caused a strong shift toward more advanced (hybrid) EMT states, both in vitro and in transplanted tumors in vivo. (10/20)

100 1

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 13/07/2026

In silico knockout simulations predicted that losing early regulators (Klf5, Pitx1) pushes cells toward more mesenchymal states, while losing late regulators (Nfatc1, Creb3l1) pulls cells back toward epithelial states. Time to test this for real. (9/20)

100 1

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 13/07/2026

This let us build a predicted TF-TF interaction network: epithelial TFs (Grhl1, Trp63, Klf5, Pitx1) form a tight circuit that gets repressed by Zeb1 as cells transition, while late-EMT TFs (Nfatc1, Ebf1, Creb3l1) get switched on downstream. (8/20)

100 1

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 13/07/2026

We then added single-cell multiome (RNA + ATAC together) to map the chromatin landscape, which enhancers open and close as cells move through EMT, and combined it with SCENIC+ to build enhancer-driven gene regulatory networks (eRegulons). (7/20)

100 1

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 13/07/2026

To find the TFs driving these transitions, we ran SCENIC (gene regulatory network inference) on the single-cell data. This recovered known regulators (Grhl2, Zeb1, Twist1...) but also flagged new candidates: Klf5, Pitx1, Nfatc1, Creb3l1. (6/20)

100 1

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 13/07/2026

Trajectory analysis (Slingshot + scFates) showed cells progress step-by-step through these states, supporting a "spectrum" model of EMT rather than a binary switch, consistent with earlier work from our lab (Pastushenko et al., Nature 2018). (5/20)

100 1

Sebastiaan (Seb) Vanuytven @svanuytven.eurosky.social · 13/07/2026

Unsupervised clustering revealed 4 discrete EMT states: Epithelial → Early hybrid → Hybrid → Late EMT. Each has its own gene signature, and we confirmed all 4 exist at the protein level by immunostaining in tumors. (4/20)

100 1