Traditional differential expression analysis (Fig. 4) only provides a snapshot of mRNA expression. We therefore employed RNA velocity analysis using the scVelo package to explore the comparative abundance of spliced and unspliced pre-mRNA transcripts in fibroblast clusters64. scVelo uses a dynamical likelihood-based model, which identifies velocity states and transcriptional dynamics of each individual cell in an unbiased manner (Supplementary Fig. 6a)64,65. Two opposing trajectories of fibroblast differentiation were identified in both the pooled and individual patient datasets, triggered either by the activation of mechanotransduction pathways or the disruption of mechanical signaling by FAK inhibition (Fig. 5a, Supplementary Fig. 7a). We found that FAK inhibition strongly increased the transcriptional activity of mechanically activated fibroblasts, resulting in a higher proportion of unspliced pre-mRNA, which accounted for 60% of all mRNA transcripts versus 30% in control and strained cells (Fig. 5b). To quantify the relationship between fibroblast clusters resulting from either mechanical activation or disruption of mechanotransduction, we applied partition-based graph abstraction (PAGA) informed by velocity-inferred directionality to quantify the relationship between fibroblast clusters resulting from either mechanical activation or disruption of mechanotransduction (Fig. 5c)66. The fibroblasts of the control group (cluster 0) were identified as the origin of the underlying Markov transition matrix, confirming their identity as root cells of fibroblast differentiation (Fig. 5d). Partition-based graph abstraction (PAGA) identified trajectory vectors pointing from the control fibroblasts toward either activated profibrotic clusters in response to mechanical strain (2, 4, 5, 7, 8) or regenerative clusters in response to mechanotransduction blockade (1, 3, 6) (Fig. 5c).
Unique molecular identifiers (UMIs) from each cell barcode were retained for all downstream analysis. Raw UMI counts were normalized with a scale factor of 10,000 UMIs per cell and subsequently natural log transformed with a pseudocount of 1 using the R package Seurat (version 3.1.1)79. Highly variable genes were identified, and cells were scaled by regression to the fraction of mitochondrial transcripts. Aggregated data were then evaluated using uniform manifold approximation and projection (UMAP) analysis over the first 15 principal components80. Automated cell-level annotations were ascribed using the SingleR toolkit (version 3.11) against the ENCODE blue database81.
Red Baron Pack Activation Code [License]l
2ff7e9595c
Comments