New Method Integrates Complex Single-Cell Data

Researchers from the University of Turku, Finland, have developed an innovative computational approach for interpreting complex single-cell data. According to the team, this new method significantly enhances the process of identifying and grouping cell types from different biological samples, a fundamental step in understanding the cellular diversity that exists in the human body. Despite the similarities between cells, no two are exactly alike. Advances in single-cell technology now permit scientists to measure thousands of molecules across thousands of individual cells simultaneously, providing valuable insights into both health and disease.

Blood samples, for example, consist of billions of red blood cells and millions of immune cells. Each cell type possesses a unique molecular 'fingerprint,' which can be distinguished through single-cell techniques combined with computational methods. When researchers handle several samples, they face the challenge of matching the same cell types between them. This essential phase, termed data integration, presents difficulties when cell types differ in abundance or are entirely absent between samples, often leading existing methods to confuse distinct cell types.

To overcome these challenges, the researchers created Coralysis, a machine learning-based algorithm capable of effectively integrating data even imbalanced data across samples. Developed by Professor Laura Elo’s Computational Biomedicine research group, Coralysis operates by progressively clustering cell identities, emulating the strategy of assembling a puzzle by grouping pieces according to features such as color, shading, and pattern. 

“We were inspired by the process of assembling a puzzle, where one begins by grouping pieces based on low- to high-level features, such as colour and shading, before looking at shape and patterns. Similarly, our algorithm progressively integrates cellular identities through multiple rounds of divisive clustering,” explains António Sousa, first author of the study published in Nucleic Acids Research.

“Coralysis provides the scientific community with a new way to study cellular diversity and gain a deeper understanding of complex single-cell data. By making it openly available, we hope to support collaboration and accelerate discoveries across the global research community,” adds Professor Elo.