Summary

Task At Hand

Targeted differentiation of hypoimmunogenic iPSCs into functional cell types and subsequent assembly into artificial tissues for organ repair and replacement holds great potential to overcome the current donor organ shortage. The main aim of the project was to find putative sub-populations within samples of iPSC-induced cardiomyocytes and EHMs. In transcriptomics, this is addressed using single cell RNA sequencing, yet, at the time of the start of thesis there was data only from bulk RNA sequencing. Here a Computational Deconvolution approach seemed capable of providing sub-population level information from bulk sequencing using a relevant single cell dataset.

Parts of thesis also focused explored the possibilty of using RNA sequencing data to identify potential microbial contaminants, as Prof. Zimmermann’s group and their work is transitioning into the clinics and as such, would need to meet the highest of standards of current Good Manufacturing Practices, a part of which deals with minimising, and checking for microbial contamination.

Work Done

• An analysis pipeline was set up to analyse the bulk RNA sequencing data from raw sequencing files to count files, from which further analysis can be easily performed.
• Possible microbial contamination was investigated using the alignment of unmapped (to human genome) reads to bacterial and viral genomes.
• An exploratory data analysis was done and global trends were established which reiterated the group’s previous findings. For example, using PCA across multiple datasets we confirmed that the cardiomyocytes within the adult heart, fetal heart, EHMs and iPSC-cardiomyocytes were globally similar and the majority of the differences between these groups are possibily based on their tissue complexity.
• Putative proportions of sub-populations within the iPSC-induced cardiomyocytes and EHMs was established using computatinal deconvolution. The findings corroborated with the current differentiation and production protocols.