Advancing Single-Cell Analysis With CellRank 2
Researchers around Prof. Fabian Theis from Helmholtz Munich and the Technical University of Munich (TUM), have developed CellRank 2, a new tool that promises to transform the way scientists study cells. This new method significantly improves our understanding of how individual cells develop and make decisions, providing deeper insights into various biological processes and diseases. The results were now published in Nature Methods.
Every cell in our body has a unique journey and destiny. Some cells become muscle cells, others turn into brain cells, and some might become part of our immune system. Understanding how these cells make these decisions is crucial for advances in medicine and biology.
Researchers led by Fabian Theis have now developed CellRank 2, a cutting-edge tool designed to map out cellular journeys: By integrating various types of biological data, the tool not only tracks where a cell is in its developmental journey but also predicts where it is headed.
“CellRank 2 represents a significant leap forward in our ability to understand and predict cellular behavior, offering unprecedented insights into the complex paths that cells follow throughout their development”, states Theis.
The Power of CellRank 2
CellRank 2 significantly enhances the accuracy, performance, and efficiency of single-cell analysis. For a cell state relevant to the development of the immune system, it identified a previously unknown cluster of putative progenitor cells that can help better understand how the immune system emerges. CellRank 2 more precisely ranks genes that likely drive cell differentiation, crucial for identifying potential medical targets. Additionally, it operates much faster, making it ideal for analyzing large datasets. The tool also offers detailed insights into how various genes control cell fate, providing a clearer understanding of the mechanisms behind cell development.
Implications for Research and Medicine
The introduction of CellRank 2 is a significant leap forward for research in many areas, including cancer, regenerative medicine, and developmental biology. By providing a clearer understanding of cell behavior, this tool could lead to the development of new therapies and improve the ability to treat various diseases.
"CellRank 2 is a game-changer for the scientific community. Its ability to integrate different types of data and provide highly accurate predictions opens new possibilities for research and medical advancements", comments Philipp Weiler, one of the two first authors of the study. Marius Lange, the other first author, adds that “version 2 of the CellRank software enables entirely new analysis workflows that will enable researchers in different biological fields to gain a deeper understanding of cell-fate decision making”.
As CellRank 2 continues to be refined and adopted, it is expected to drive significant advancements in the understanding of cellular processes and disease mechanisms, ultimately contributing to better healthcare solutions.
Helmholtz Software Award 2024
The team around Fabian Theis, Philipp Weiler (both Helmholtz Munich) and Marius Lange (formerly Helmholtz Munich, now ETH Zürich), has won the Scientific Originality Prize (EUR 5,000) in the inaugural Helmholtz Software Award for their work on CellRank. CellRank, a tool for analyzing cell state dynamics and fate decisions using single-cell data, was chosen from 42 proposals evaluated by international experts. The award was announced by Helmholtz President Prof. Dr. Otmar D. Wiestler on July 15, 2024.
About the scientists
Prof. Dr. Fabian Theis, Head of the Computational Health Center at Helmholtz Munich and Chair of the Mathematical Modeling of Biological Systems at the Technical University of Munich (TUM)
Philipp Weiler, PhD candidate at the Computational Health Center at Helmholtz Munich
Marius Lange, Postdoctoral Researcher at the Department of Biosystems Science and Engineering of ETH Zürich
Original publication
Weiler et al. (2024): CellRank 2: unified fate mapping in Multiview single-cell data. Nature Methods. DOI: 10.1038/s41592-024-02303-9
