A new computational method correctly distinguishes between data from tumor cells and normal cells

In an effort to address a key challenge when analyzing large single-cell RNA-sequence datasets, researchers from the University of Texas MD Anderson Cancer Center have developed a new computational technique to accurately differentiate make a comparison between data from cancer cells and the different normal cells found within tumor samples. The work was published today in Nature’s biotechnology.

The new tool, called CopyKAT (karyotyping copy of aneuploid tumors), allows researchers to more easily analyze the complex data obtained from large unilateral RNA experiments, which provide gene expression data from thousands of individual cells.

CopyKAT uses that gene expression data to look for aneuploidy, or the presence of rare chromosome numbers, which are common in most cancers, said study lead author Nicholas Navin, Ph.D. ., associate professor of Genetics and Bioinformatics & Computational Biology. The tool also helps identify specific subunits, or clones, within the cancer cells.

We developed CopyKAT as a tool to extract genetic information from the transcription data. By applying this tool to several datasets, we showed that, with approximately 99% accuracy, we were able to unrecognize tumor cells against the other immune or stromal cells present in mixed tumor sample. We could then go a step further to discover the underlying stones that are present and understand their genetic differences. “

Nicholas Navin, Ph.D., Associate Professor of Genetics and Bioinformatics & Computational Biology

Historically, tumors have been studied as a combination of all cells present, many of which are non-cancerous. The advent of single-cell RNA sequence in recent years has allowed researchers to study tumors in a much larger solution, examining the gene expression of each individual cell to develop a picture of the tumor landscape, introduction of the surrounding micro-environment.

However, it is not easy to differentiate between cancer cells and normal cells without a reliable computational method, Navin explained. Former postdoctoral fellow Ruli Gao, Ph.D., now an associate professor of Cardiovascular Sciences at the Houston Methodist Research Institute, developed the CopyKAT algorithms, which develop ancient methods by ‘increasing accuracy and adapting for the latest generation of single-cell RNA data. .

The team first benchmarked their instrument by comparing results with full-genome sequence data, which showed high accuracy in predicting changes in copy number. In three additional data sets from pancreatic cancer, triple-negative breast cancer and anaplastic thyroid cancer, the researchers showed that CopyKAT was inaccurate in differentiating between tumor cells and normal cells in mixed samples.

These analyzes were made possible through collaboration with Stephen Y. Lai, MD, Ph.D., professor of head and neck surgery, as well as Stacy Moulder, MD, professor of Breast Medical Oncology, and the Breast Cancer Moon Shot^®, part of MD Anderson’s Moon Shots Program®, a collaborative effort to rapidly advance scientific discoveries to meaningful clinical advances that will save patients ’lives.

In analyzing these samples, the researchers also showed that the tool is effective in identifying subpopulations of cancer cells within the tumor based on copy number differences, such as confirmed by experiments in triple-negative breast cancers.

“Using CopyKAT, we were able to identify rare subtypes within triple-negative breast cancers that have specific genetic changes that have not been widely reported, including those with therapeutic effects. that could be there, “Gao said. “We hope this tool will be useful to the research community to make the most of their RNA single-cell sequence data and to manage new findings in cancer.”

The tool is available free to researchers here. The authors note that the tool is not relevant to the study of all types of cancer. Aneuploidy, for example, is very rare in pediatric and hematologic cancers.