Vanessa Dumeaux 🇨🇦
banner
vdumeaux.bsky.social
Vanessa Dumeaux 🇨🇦
@vdumeaux.bsky.social
Assistant Professor, Anatomy & Cell Biology, Western University, Canada. Computational Biology, Genomics, Microbes and Human Health https://lab-dumeaux.science/
very cool paper: t.co/mxObUMC8R4 Cancer cells can transfer their mitochondria to T-cells which impairs the immune response t.co/dkEv5L5Bb0
October 31, 2025 at 8:26 PM
10/11 Copy number changes tell another interesting story. We found: • Subtype-specific patterns (basal-like = high CNA burden) • 6 regions linked to recurrence risk • Key changes in tissue architecture genes (cadherins) This adds another layer to our understanding of DCIS progression! 🧬
March 6, 2025 at 4:36 PM
9/11 We are the first to identify specific genetic markers of radiotherapy response in pure DCIS! These are co-occurring mutations affecting tissue architecture and cell-matrix interactions in radiotherapy-treated patients. Not related to tumor mutational burden 🎯
March 6, 2025 at 4:36 PM
8/11 We identified 12 mutually exclusive genes linked to early recurrence risk, operating in three major pathways: • Cytoskeleton & vesicle dynamics • Signal transduction & receptor activity • DNA/RNA regulation
March 6, 2025 at 4:36 PM
6/11 Key mutation patterns: PIK3CA leads (15%), but the real story? Widespread mutations in genes controlling tissue architecture: Motor proteins (DNAH12, DNHD1), Collagen genes (COL18A1, COL4A3), Mucins (MUC3A, MUC4) These suggest disrupted cell adhesion & tissue structure are central to DCIS! 🧫
March 6, 2025 at 4:36 PM
5/11 We found huge variation (3-3,482 variants/sample). A subset of younger patients (<50y) showed distinct COSMIC signatures with higher mutation loads highlight distinct mutational processes in some early-onset DCIS cases. Yet this is not associated with prognosis and response to therapy.
March 6, 2025 at 4:36 PM
4/11 Technical highlight: We used Neusomatic, a deep learning approach combining 6 different variant callers + matched normal data to identify mutations. Why? Because 88.7% of variants were only caught by single callers! Our method ensures robust variant detection even in challenging FFPE samples.🤖
March 6, 2025 at 4:36 PM
🦠 Disease & Functional Confounding
* #IBD microbiomes often shift toward Archetype 2 linked to inflammation
* Archetype 1-dominant IBD samples show increased carbohydrate metabolism
* Archetype 3-dominant IBD samples are enriched in pathways linked to E. coli adherence and immune activation
8/n
January 31, 2025 at 3:31 PM
While specific associations exist between functional archetypes and compositional enterotypes, the relationship is not strictly deterministic.
7/n
January 31, 2025 at 3:31 PM
🔬 What are these archetypes?
*Archetype 1: High carbohydrate metabolism → BCAA & microbial cell wall biosynthesis
*Archetype 2: Fatty acid metabolism + TCA cycle → highest metabolic flexibility = most stable & diverse
*Archetype 3: Amino acid + nitrogen metabolism → linked to inflammatory pathways
January 31, 2025 at 3:31 PM