News & Events
News
03 October 2025
Unlocking Hidden Cells: A New Tool Reveals What’s Really Inside a Tumor
xCell 2.0 is a versatile and robust tool that maintains high performance across various reference types and biological contexts
Technion researchers create an algorithm that sheds new light on tumor composition.
A research team led by RTICC member Assistant Prof. Dvir Aran and doctoral student Almog Angel from the Faculty of Biology at the Technion – Israel Institute of Technology, has introduced xCell 2.0, a next-generation computational tool that can accurately estimate the proportions of different cell types within complex tissue samples. Recently published in Genome Biology, this innovation enables researchers to extract detailed biological insights from existing gene-expression data, making it particularly valuable for cancer studies.
Making Sense of Complex Tissues
Tumor samples are mixtures of many cell types — not only cancer cells but also immune cells, supporting stromal cells, and blood vessel cells. Understanding this complex mix is crucial because the cellular environment around a tumor, known as the tumor microenvironment (TME), plays a major role in how cancers grow and respond to therapy. Traditionally, such analysis required expensive single-cell technologies. With xCell 2.0, researchers can use more common bulk gene-expression data to reveal these details quickly, accurately, and at a fraction of the cost.
Smarter, More Accurate, and Built for Cancer Research
The Technion team enhanced the algorithm to recognize unique “signatures” of each cell type and to correct for overlaps between similar cells. Tested on dozens of datasets, xCell 2.0 outperformed existing methods, delivering more reliable cell-type estimates across a range of tissues. For cancer research, this means scientists can now better assess how immune cells infiltrate tumors — a key factor that influences patient response to treatments such as immunotherapy.
In a test example of pan-cancer immune cell checkpoint blockade response prediction, xCell 2.0-derived TME features significantly improved prediction accuracy compared to models using only cancer type and treatment information. The method also outperformed other computational deconvolution tools and established prediction scores, demonstrating its strong potential to guide precision oncology.
The Research has been published in Genome Biology
