Journal: Nature
Article Title: Epigenetic memory of colitis promotes tumour growth
doi: 10.1038/s41586-026-10258-4
Figure Lengend Snippet: a, Fluorescent in situ hybridization against GFP for clonal barcode transcripts in HEK293T cells. Top, cells infected with original LARRY barcode from Weinreb et al. Bottom, cells infected with LARRY barcode modified by addition of mU1 hairpin sequence. Scale bar 50 μm. Representative of n = 100 cells across 2 independent experiments. b, Schematic detailing molecular process of capturing clonal barcodes and computational workflow for assigning cells to clones. c, HNF4/PPAR motif accessibility amongst the 50 largest clones by cell number. Each violin plot represents distribution of motif scores for an individual clone and point represents median values per clone. d, Framework for computing clonal variance and identifying clonal features. e, Motif accessibility relative to position in the colon. The X-axis represents segment of the colon, with segment 1 being most proximal and segment 6 being most distal. The Y-axis shows change in motif accessibility of a given segment relative to segment 1. f, Accessibility of all TF motif families relative to position in the colon. Each row represents a TF motif family and color indicates change in motif accessibility relative to segment 1. g, Expression stemness ( Lgr5 , Lrig1 ) and differentiation ( Car1 ) markers in relation to HNF4/PPAR motif score. h, Downsampling analysis of permutation-based clonal heritability. Clones were randomly sampled to the number indicated on the x-axis and permutation-based testing was performed identically to the complete dataset. The total number of motif families identified to have lower clonal variance as compared to random (FDR < 0.05) for each clone number is shown on the y-axis. i, Linear mixed model of random effects from exposure to colitis and clonal identity. j, Variance explained by clonal identity from the linear mixed model. k, Comparison of variance explained by clonal identity and exposure to colitis for all TF motif families. l, Left, variance explained by clonal identity and exposure to colitis for top TF motif families. Right, variance attributed to each factor following randomization of clonal identities within colitis or control conditions. m, Comparison of clonality identified by permutation and linear mixed model. The X-axis shows -log10(FDR) from the permutation method (Fig. ) and the Y-axis shows percent variance attributed to clonal identity in the linear mixed model (panel j). Panels created in BioRender: e , Nagaraja, S. https://biorender.com/jqhj3wt (2026); i , Nagaraja, S. https://biorender.com/4mut8sd (2026).
Article Snippet: Human embryonic kidney 293T (HEK293T) cells (American Type Culture Collection (ATCC), CRL-3216; authenticated by short tandem repeat profiling and tested for mycoplasma by ATCC) were grown in DMEM (Thermo, 11965-092) with 10% fetal bovine serum (FBS) and 1% penicillin–streptomycin.
Techniques: In Situ Hybridization, Infection, Modification, Sequencing, Clone Assay, Expressing, Comparison, Control