Revolutionizing Morphological Cell Profiling with IKOSA's Innovative Cell Painting Assay

Have you ever wondered how the intricate dance of molecules within our cells can be unraveled to better understand disease mechanisms and develop new treatments??

The world of cellular biology is undergoing a revolutionary transformation, and at the forefront of this evolution is IKOSA's groundbreaking Cell Painting Assay. ??

In this edition of our newsletter, we delve into the incredible advancements brought about by IKOSA's innovative technology, which is poised to reshape the way we study and analyze cellular structures and functions. Join us as we explore the fascinating world ?? of morphological cell profiling and how IKOSA is leading the charge towards a deeper understanding of cellular biology. ??

The Cell Painting Method and Morphological Cell Profiling explained.

The Cell Painting method begins with the labeling of cells using fluorescent dyes that bind selectively to various cellular structures, including the nucleus, cytoplasm, and specific organelles. These dyes emit unique fluorescent signals, allowing for the simultaneous visualization of multiple cellular components. Subsequently, fluorescence microscopy captures detailed images of these labeled structures.?

Morphological cell profiling is the systematic study of cellular shape, size, texture, and spatial arrangement. It relies on advanced imaging, like high-resolution microscopy, to collect detailed data on cellular features. This data is then used to quantify the morphology of individual cells or cell groups.

How did we do it?

We introduce an AI-powered approach for analyzing Cell Painting image data, utilizing the forthcoming IKOSA software. This App, designed for morphological cell profiling, is the result of our collaboration with Core Life Analytics , a Dutch company specializing in downstream analytics tools for high-content data. The App is currently in its final development phase.

???? For a deeper dive into this project, head over to our scientific poster that we presented at SLAS Belgium in May 2023.

The App Training Results

The trained model’s performance is assessed through a comprehensive evaluation of its segmentation output, involving both qualitative and quantitative measures. Additionally, it is compared to the state-of-the-art JUMP CP pilot dataset for a thorough analysis of its performance.

The model is trained using 8 imaging channels and over 2,200 images with labels. The?training data?comprises over?215,732 nucleus labels?and?231,501 cell labels. The?validation data?consists of?572 images?with?58,290 nucleus annotations?and?62,560 cell annotations. It demonstrates efficient performance by making predictions for 1,080x1,080 pixel-sized images in just 2.2 seconds. The quantitative evaluation shows high accuracy in instance identification and a low rate of False-Positives, highlighting its effectiveness.

In the qualitative evaluation, our model’s nucleus segmentation outperforms CellProflier’s ground truth data, showcasing precise and accurate segmentation, even in challenging situations like cell accumulation.

Benefits of the IKOSA Cell Painting APP

• User-friendly
• Fast and automated
• Robust and High Performance
• Transferable and Reusable
• Collaboration with Core Life Analytics?
• Resource Efficient
• Dedicated Support and Maintenance?

???? Read more about our experience creating this use case in our extended blog post here:?A New Way to Morphological Cell Profiling Using IKOSA’s Cell Painting Assay.

The upcoming IKOSA Cell Painting App aims to set a new standard for robust and versatile solutions in the field. Our goal is to offer users a tool that significantly advances drug discovery and personalized medicine through comprehensive morphological profiling.

Your Inside AI Newsletter Team?????

Thanks for reading this article! Let us know in the comments below what you think about this use case and be sure to subscribe to our biweekly Inside AI in Life Sciences newsletter, to get the latest tips, tricks, and news. Also, follow us here on?LinkedIn.