The field of Advanced Therapy Medicinal Products (ATMPs) and precision medicine is rapidly advancing, with cell therapies based on induced pluripotent stem cells (iPSCs) showing particular promise in treating a range of diseases, such as off-the-shelf immune therapy for cancer (e.g. CAR T and CAR NK cells), diabetes, and spinal cord injury, myocardial infarction, Parkinson’s disease (PD), and age-related macular degeneration. With their unlimited potential to expand and to create any cell type in the body, combined with the long-term storage of cell therapy doses, iPSCs are the ultimate starting material for off-the-shelf cell therapy for many indications. Generation of commercially available iPSC-derived cell therapy for precision medicine is currently a costly, complex, and advanced process; the full potential of iPSCs as a starting material is not yet fully utilised. Not only are the processes of expanding and quality testing the iPSCs expensive, but they are also close to impossible to perform with current technologies. Therefore, technology development is needed to: i) scale up and introduce free-floating culturing, ii) develop minimal recruitment cell culture media and matrices, iii) create novel quality controls using machine learning, and iv) replace costly, unethical, and labour-intensive animal testing with AI models. One of the main objectives of StartCell is to develop methods for quality control and release criteria of iPSC start material. These methods will be based on gene expression patterns and cell morphology characteristics, to replace as much of the animal testing as possible, and to develop cost-effective alternatives to quality control. Different goals are required to achieve this objective: 4.1 Generation of single-cell RNA sequencing (scRNA-seq) data and microscopic images of iPSC: -Preprocessing of the scRNAseq data and microscopic images for machine learning consumption. -Development of a Feed Forward Neural Network (FFNN) using a pretrained FFNN model developed for the classification of embryonic stem cells (ESC) from scRNA-seq data that will be available at project start. The pretrained model will be finetuned on scRNA-seq data from iPSCs generated in this project to enable the classification based on cell quality. 4.2 Interpretation of the developed scRNA-seq-based FFNN model for the identification of novel genes as candidate biomarkers for the implementation of quality control (QC) of iPSC in industrial settings. Validation of identified predictive genes as candidate QC biomarkers using additional iPSC lines. 4.3 Development of a CNN classification model based on cell morphology using microscopic images from iPSC cultures of high and low quality for QC implementation in large-scale production. 4.4 Developing a prototype for a QC decision support system for iPSC with release criteria based on gene expression and cell morphology.