Journal: bioRxiv
Article Title: Scalable longitudinal imaging and transcriptomics of cells in dynamic enclosures
doi: 10.64898/2026.05.05.723030
Figure Lengend Snippet: A: classification of CCEs in different phenotypes based on the analysis of longitudinal imaging data. Red: CellTrace™ Far Red, blue: Annexin V, green: EGFR. B: UMAP based on the transcriptomic data from 10,604 CCEs containing A549 cells treated with 10 µM Olmutinib. The colors represent different transcriptomic clusters. C: UMAP based on the transcriptomic data (same as panel B) colored according to the imaging-derived CCE classification in panel A. 2,328 CCEs that could not be accurately classified were excluded from the analysis. D: proportion of CCEs (y axis) belonging to each imaging-based phenotype (indicated by the color) within each gene expression cluster (x axis). E: Upset plot showing overlap of significant GSEA pathway enrichments across three classification strategies. The combination of transcriptomic clustering with imaging classification identified 15 unique pathways not found in either single-modality strategy. F: significant interaction effects (p_adj < 0.05) between RNA clusters and imaging phenotypes on the prediction of drug resistance pathway modules (G2M checkpoint, E2F targets, MYC targets, DNA repair, EMT) (see Methods). The daughter cell resistant phenotype showed 7 out of 14 total significant interactions, indicating that pathway activities are maximally explained by the combination of transcript state and the daughter cell resistant phenotypic classification. G: Confusion matrix for elastic net prediction of imaging phenotypes from gene expression. H: STRING PPI network for top 50 positive coefficient genes (associated with daughter cell resistance). I: STRING PPI network for top 50 negative coefficient genes (inversely associated with daughter cell resistance). J: Selected differentially expressed genes between expression-defined clusters (x axis). The color represents the average expression (scaled per gene) and the size of the circle indicates the percentage of CCEs expressing the gene. Cluster 2 showed strong enrichment for cell division pathways and overexpressed the proliferation marker TOP2A. Cluster 3 exhibited activation of multiple EGFR bypass pathways with overexpression of EPHA7 (64), HGF (65), ERBB2 (66), and AXL(67), all capable of activating MAPK signaling independently of EGFR. Cluster 5 displayed enrichment of p53 targets, including upregulation of quiescence-associated genes such as GADD45A, REDD1, ATF3, SFN, and BTG2.
Article Snippet: Human lung cancer A549 cells were purchased from ATCC (CCL-185), and cultured in DMEM supplemented with 10% FBS and 1% Pen-Strep.
Techniques: Imaging, Derivative Assay, Gene Expression, Expressing, Marker, Activation Assay, Over Expression
ATCC is a verified supplier 
ATCC manufactures this product 
