Donor selection metrics for different donors that were evaluated for the study

. Donor selection metrics for different donors that were evaluated for the study.

Donor selection, myotube quantification and dystrophin expression validation. (A) Primary myoblasts from DMD and non-DMD donors were selected among other donors based on retaining their capacity to differentiate (% Desmin+ myoblasts and Fusion Index) into myotubes within the MyoScreen platform. Cells expanded following patient biopsy collection were subsequently enriched for myoblasts using flow cytometry (CD56+). Primary vials were sourced, thawed, and the proportion of Desmin+ cells was determined using flow cytometry. Cells were expanded and cryopreserved into master banks (MB) at which point they were characterized using flow cytometry (Desmin+ cells) and the MyoScreen platform (fusion index). Finally, master bank vials were thawed, expanded, and finally cryopreserved into working cell banks (WB) at which point they were characterized using flow cytometry (Desmin+ cells) and the MyoScreen platform (fusion index). Cells were selected based on consistency in their doubling time, proportion of Desmin+ cells and fusion index. (B) Myoblasts from the two non-DMD (Non- DMD #1, Non-DMD #3) and two DMD (DMD #1, DMD #4) donors were differentiated for 10 days in MyoScreen micropatterned plates. The differentiated myotubes were stained for Myosin Heavy Chain (MHC) and Dystrophin protein. Sample images of individual Myoscreen islands for each donor stained for Myosin heavy chain (MHC). (C) Quantification of nuclei count, myotube area and fusion index for myotubes generated from the 4 different donors. Overall, the fusion index which quantifies muscle cell myotube formation, is significantly lower in DMD donor cells compared to non-DMD donors. (D) Sample fluorescent images from the 4 donors stained for dystrophin. Dystrophin was not observed with this endpoint in differentiated myotubes from DMD donors. (E) Quantification of dystrophin intensity from DMD and non-DMD donors. (F) Dystrophin protein expression was further evaluated using Western blotting.

Cell profiling of myotubes using only the individual (Utrophin and α-Sarcoglycan) DAPC marker features. (A) t-SNE plots of multi-dimensional cell profiler features. The left panel is derived using only the utrophin channel features while the right panel is derived using the α- Sarcoglycan features. Each point is a single myotube and the color indicates the donors. Utrophin displays a higher separation on the t-SNE plot compared to α-Sarcoglycan. (B) Support vector Machine (SVM) 10-fold cross validation data confirms higher separation with Utrophin features compared to α-Sarcoglycan. Further the coefficient of variation (CV) is lower for Utrophin. (C) Sample images of fluorescent labelled images for DAPC proteins utrophin and α-Sarcoglycan for Non-DMD #1 (non-DMD) and DMD #1 (DMD) donors. There was no difference between average intensity of utrophin and α-Sarcoglycan between the DMD and non-DMD donors (Supplementary Fig S2.1).

DAPC protein combinations display superior performance when used in combination with Utrophin. (A) Cross validation F-Score from morphological features of other DAPC proteins in combination with utrophin from cells cultured on MyoScreen platform. Combinations of DAPC proteins with Utrophin achieve better performance in separating the DMD and the non-DMD donor population when compared to individual DAPC proteins. Utrophin in combination with α- Sarcoglycan displayed much higher performance (F-Score .99) compared to either single DAPC protein alone or other DAPC protein combinations with Utrophin. (B) t-SNE plots showing donor cells from the MyoScreen platform for DAPC protein combinations. Comparing the t-SNE plots, the high separation between the DMD and non-DMD cells is highly evident in “Combination-1” (Utrophin + a-Sarcoglycan) compared to other DAPC protein combinations with utrophin. (C) Cross validation using the subcellular features using DAPC proteins in combination with α- Sarcoglycan obtained from cell profiling from cells cultured on the MyoScreen platform. Combinations of DAPC proteins with α-Sarcoglycan also shows better performance in separating the DMD and the non-DMD donor population when compared to individual DAPC alone. (D) t- SNE plots showing donor cells from the MyoScreen platform for DAPC protein combinations.

Quantification of functional relevance of Dystrophin knock down (KD) using cell profiling. (A) siRNA-based knock down of Dystrophin in non-DMD differentiated myotube cells (Non-DMD #1 and Non-DMD #3). The image intensity reflects dystrophin expression. For comparison, Dystrophin expression from the DMD-donors is included on the right. This is a baseline level of Dystrophin expression. In comparison, we saw that Dystrophin expression using siRNA3 is comparable to Dystrophin expression detected from the DMD donors. All images are pseudo colored. (B) F-score from cross validation analysis using SVM linear kernel by taking the 2 different categories at a time using Utrophin and α-Sarcoglycan features. As evident from the F-score, we can separate siRNA3 samples from both DMD and non-DMD scramble siRNA. (C) t-SNE plot of Utrophin and α-Sarcoglycan features. Each dot is a myotube within a pattern. Shades of green indicates scramble siRNA treatment for non-DMD donors while shades of red indicate scramble siRNA treatment for the DMD donor cells. The blue dots indicate the siRNA3 dystrophin knock down of the non-DMD donor cells. (D) Quantification by population Euclidean distance. Euclidean distances were measured using centroid of different populations. (E) Using an SVM classifier trained on the scramble siRNA between the DMD and the non-DMD donors, we predicted the siRNA dystrophin KD myotubes. The y-axis shows the prediction of the classifier as non-DMD like myotubes. As evident, by knocking down dystrophin in the non-DMD donor cells, we can generate a phenotype that similar the DMD phenotype (% Like DMD Non-DMD #3 94.8%, Non-DMD #1 – 80.2%).

- Quantification of functional relevance of Dystrophin by exon skipping. (A) DMD donor DMD #1 was chosen for performing exon skipping and restoring partial Dystrophin expression. Images of non-DMD (Non-DMD #3) untreated as well as exon 45 targeting in DMD #1 are displayed. Partial restoration of Dystrophin is observed after exon 45 skipping. All images are pseudo colored. (B) F-score from cross validation analysis using SVM linear kernel by taking the 2 different categories at a time using Utrophin and α-Sarcoglycan features. As evident from the F-score, we can separate exon 45 skipped samples from non-DMD controls. (C) t-SNE plot of Utrophin and α-Sarcoglycan features. Each dot is a myotube within a pattern. Shades of green indicates non-DMD donor while shades of red indicate untreated DMD donor cells. The blue dots indicate the partially restored dystrophin from DMD #1 donor cells after exon skipping. (D) Quantification by population Euclidean distance. Euclidean distances were measured using centroid of different populations. (E) Using an SVM classifier trained on the DMD and the non- DMD donors. The y-axis shows the prediction of the classifier as non-DMD like myotubes. As evident, after partial restoration of dystrophin expression in the DMD donor cells, 41.3% of the cells are classified as non-DMD phenotype.

Primary antibodies

Quantification of DAPC protein expression across different donors. (A-G) Average pixel intensity of individual DAPC protein (a measure of DAPC protein expression level) across multiple myotubes of each DAPC protein. Significant differences in signal intensity were not detected for any of the DAPC proteins analyzed.

Cross validation accuracy for other DPAC proteins excluding Utrophin and a- Sarcoglycan. (A) Cross validation using the subcellular features using individual DAPC proteins obtained from cell profiling from cells cultured on MyoScreen platform. The corresponding t-SNE plates are displayed in the different plots.

Validation of the MyoScreen profiling platform. (A) Fusion index of myotubes cultured using standard 96 well plates. No significant differences between DMD and non-DMD donors were observed. (B) Cross validation using the subcellular features obtained from cell profiling from cells cultured in standard 96 well plates compared with cells cultured in the MyoScreen platform. A significantly higher cross validation F-Score can be observed when cells are cultured in the MyoScreen platform.