In this interview, we speak with Angeline Lim, PhD, Senior Application Scientist at Molecular Devices, about the role of AI and automation in modern drug discovery. Angeline shares how Molecular Devices is transforming high-content imaging and cell culture with their AI-enabled systems, including the CellXpress.ai™ Automated Cell Culture System, which uses machine learning to streamline complex workflows and make assays more reliable and reproducible.
Could you please introduce yourself and describe your role at Molecular Devices?
My name is Angeline Lim, and I’m a Senior Application Scientist at Molecular Devices. I’m responsible for developing and testing new workflows and applications on our high-content imaging systems. I work with high-content imaging, 3D biology, image analysis, and automation. Over the last few years, we’ve begun exploring how AI can solve complex image analysis problems for our customers.
Molecular Devices is one of the world’s leading providers of high-performance bioanalytical measurement systems, software and consumables for life science research, pharmaceutical and biotherapeutic development. Included within a broad product portfolio are platforms for high-throughput screening, genomic and cellular analysis, colony selection, and microplate detection. These leading-edge products enable scientists to improve productivity and effectiveness, ultimately accelerating research and the discovery of new therapeutics. Molecular Devices is committed to the continual development of innovative solutions for life science applications. The company is headquartered in Silicon Valley, California, with offices around the globe. For more information, please visit www.moleculardevices.com.
AI is revolutionizing all areas of scientific research. How is Molecular Devices utilizing AI to enhance its workflows and support its customers?
At Molecular Devices, we’re passionate about equipping researchers with next-generation technology that advances scientific discovery. We recognize that AI is revolutionizing biological research, and we’re creating solutions that allow our customers to leverage the power of AI in their own laboratories.
For high-content imaging specifically, AI helps meet the challenges of complex image analysis. Our IN Carta® Image Analysis Software uses AI to uncover and analyze complex phenotypic data. For example, it features a deep-learning-based segmentation tool that can analyze complex or traditionally hard-to-analyze images, such as those acquired with brightfield imaging.
Users don’t need any coding or scripting experience to get started with our IN Carta analysis software. It’s intuitive and easy to use – almost like a coloring book. You simply mark the objects you’re interested in (a process known as annotation) to train the AI and then test the model. If the model works well, you save and use it in your image analysis protocol. If it doesn’t, you can retrain it by adding more annotated images. It’s an iterative process with human supervision.
This software is also integrated into our CellXpress.aiTM Automated Cell Culture System.
The CellXpress.ai system is an enclosed, easy-to-use solution for cell culture that enables our customers to perform simple and complex cell culture protocols without the need for advanced expertise in scripting or automation.
By using AI, the CellXpress.ai system enables users to standardize cell culture processes, replacing previously subjective decisions with objective measurements that determine the next stage of the cell culture process, such as the best time to passage organoids or stem cells. By removing subjective assessments that typically differ between individuals, the CellXpress.ai system improves the reproducibility of cell cultures while still allowing manual input if needed.
With automation comes the need for precision. How is the technology trained to avoid errors?
A key benefit of our technology is that it reduces human errors. The system has safeguards in place; before starting an experiment, the software checks for essentials like sufficient media. If something is missing, the software prompts the user to address any gaps before the experiment runs. The system will also prompt you if it predicts an error will occur as a result of changing elements within the process. The system can also send alerts via email to inform users of potential errors or simply notify them when an automated decision is made. The system provides traceability, allowing users to track down the source of errors, unusual results, or contamination.
Could you give example workflows for the CellXpress.ai Automated Cell Culture System?
Before launch, we developed pre-configured protocols to save customers time on optimization, especially liquid handling.
We validated three main workflows, including one for iPSC culture, which is notoriously high-maintenance and requires daily media changes. The protocols for these workflows are optimized to ensure that critical elements, such as media changes and passaging, occur automatically. By reducing this manual workload, scientists don’t have to come into the lab on weekends to maintain their cell cultures.
Another workflow uses spheroids, where we fine-tuned the media exchange to avoid damaging or losing spheroids. Finally, the third was for intestinal organoids, where we carried out the entire process of 3D cell culture from seeding and feeding to passaging. Using AI and automation here addresses key reproducibility challenges within 3D biology.
Image Credit: Molecular Devices
What impact do you see the CellXpress.ai Automated Cell Culture System having on researchers?
I think it will change their lives. I’ve spoken to scientists who wished they had access to this system when they were in grad school. They can still complete their work but get their weekends back while focusing more on their science rather than the necessary but mundane manual tasks.
Automation and AI are crucial in early-stage drug discovery. What do you hope your clients can achieve?
I hope they get answers faster.
By catching drug failures early, researchers can save a lot of time and money. Using automation and AI with advanced cellular models helps better predict which compounds will work, so we don’t have to rely as much on expensive animal models. This way, fewer drugs fail during clinical testing, making the development process smoother and more efficient.
About Angeline Lim
Dr. Angeline Lim is a Sr. Applications Scientist at Molecular Devices, where she specializes in automation, high-content imaging, and analysis. She works predominately in developing automated 3D biology workflows and leverages AI to derive actionable readouts from complex phenotypic assays. She provides scientific support for the company’s portfolio of ImageXpress® High-Content Imaging Systems and their applications in high-throughput phenotypic profiling and 3D models in biology. Dr. Lim has several patents for her work in the auto
mation of 3D cell culture and has been published in various scientific publications. In addition to her work at Molecular Devices, she is an active member of the SBI2 board and runs one of the educational courses at SLAS. Dr. Lim has over 10 years of research experience and holds a PhD in Molecular Cell and Developmental Biology from the University of California at Santa Cruz.