A live-cell image-based machine learning strategy for reducing variability in PSC differentiation systems

by Selleckchem | Dec 09 2023

The differentiation of pluripotent stem cells (PSCs) into diverse functional cell types provides a promising solution to support drug discovery, disease modeling, and regenerative medicine. However, functional cell differentiation is currently limited by the substantial line-to-line and batch-to-batch variabilities, which severely impede the progress of scientific research and the manufacturing of cell products. For instance, PSC-to-cardiomyocyte (CM) differentiation is vulnerable to inappropriate doses of CHIR99021 (CHIR) that are applied in the initial stage of mesoderm differentiation. Here, by harnessing live-cell bright-field imaging and machine learning (ML), we realize real-time cell recognition in the entire differentiation process, e.g., CMs, cardiac progenitor cells (CPCs), PSC clones, and even misdifferentiated cells. This enables non-invasive prediction of differentiation efficiency, purification of ML-recognized CMs and CPCs for reducing cell contamination, early assessment of the CHIR dose for correcting the misdifferentiation trajectory, and evaluation of initial PSC colonies for controlling the start point of differentiation, all of which provide a more invulnerable differentiation method with resistance to variability. Moreover, with the established ML models as a readout for the chemical screen, we identify a CDK8 inhibitor that can further improve the cell resistance to the overdose of CHIR. Together, this study indicates that artificial intelligence is able to guide and iteratively optimize PSC differentiation to achieve consistently high efficiency across cell lines and batches, providing a better understanding and rational modulation of the differentiation process for functional cell manufacturing in biomedical applications.

Comments:

The passage you've provided discusses a significant challenge in stem cell research and highlights a novel approach to addressing this challenge using a combination of live-cell bright-field imaging and machine learning (ML) techniques. Here's a breakdown of the key points in the passage:

### Background:
Pluripotent stem cells (PSCs) have the potential to differentiate into various functional cell types, making them valuable for drug discovery, disease modeling, and regenerative medicine. However, differentiation processes are hindered by substantial variabilities between different cell lines and batches. For example, in the differentiation of PSCs into cardiomyocytes (CMs), inappropriate doses of certain chemicals can lead to misdifferentiation.

### Approach:
1. **Live-Cell Bright-Field Imaging:** Utilizing live-cell bright-field imaging techniques to monitor the entire differentiation process in real-time.

2. **Machine Learning (ML) Integration:** Integrating machine learning algorithms to recognize different cell types (CMs, cardiac progenitor cells, PSC clones, and misdifferentiated cells) in real-time. This ML approach allows for non-invasive prediction of differentiation efficiency.

### Benefits and Applications:
1. **Predictive Differentiation Efficiency:** ML enables the prediction of differentiation efficiency in real-time, allowing researchers to assess the progress and quality of differentiation.

2. **Cell Purification:** ML-recognized CMs and cardiac progenitor cells can be purified, reducing contamination and ensuring a purer cell population.

3. **Optimizing Chemical Doses:** ML facilitates early assessment of chemical doses (e.g., CHIR99021) to correct misdifferentiation trajectories, ensuring the correct development of cells.

4. **Start Point Control:** Evaluation of initial PSC colonies using ML helps in choosing the optimal starting point for differentiation, enhancing the efficiency of the process.

5. **Chemical Screening:** ML models are employed for chemical screening, leading to the identification of a CDK8 inhibitor that improves cell resistance to chemical overdoses.

### Significance:
1. **Consistent Differentiation:** ML-guided differentiation provides a method that is more resistant to variabilities, ensuring consistent efficiency across different cell lines and batches.

2. **Biomedical Applications:** The study suggests that this approach not only enhances scientific understanding but also facilitates the rational modulation of differentiation processes, crucial for functional cell manufacturing in biomedical applications.

In summary, the integration of live-cell imaging and machine learning techniques offers a promising solution to the challenges posed by variabilities in stem cell differentiation. It enables real-time monitoring, predictive analysis, and optimization of the differentiation process, leading to more reliable outcomes in biomedical applications.