Tumors are not made up of just one type of cell; instead, they consist of a mix of cells that vary genetically and functionally. This diversity in tumor cells, known as heterogeneity, is a key factor in why cancer treatments can sometimes fail, especially when cells evolve to resist drugs. Understanding this complexity can help researchers develop more effective therapies.

One way to study the diversity of tumor cells is by creating a phylogenetic tree. This tree shows how different tumor cells have evolved over time, with each branch representing a different population of cells that has acquired mutations. To create this tree and understand the mutation patterns in tumors, scientists need sophisticated tools.

Researchers at the University of North Carolina at Chapel Hill have developed a new tool called Canopy2 to help researchers do just that. Canopy2 is a Bayesian framework designed to use genetic data, specifically single nucleotide variants (SNVs) from both bulk DNA and single-cell RNA sequencing, to create a detailed view of the evolutionary history of tumor cells. The framework uses complex statistical methods (like Markov chain Monte Carlo) to account for the challenges of single-cell sequencing data, which often has gaps or “zeros” due to various factors, like mutations not being expressed in some cells or not detected by sequencing technology.

Illustration of tumor phylogeny inference through separate analysis with bulk DNA-seq and scRNA-seq

(A) Cancer phylogenetic tree representing the ground truth with six single nucleotide variants (SNVs). The leftmost, non-bifurcating branch denotes the population of normal cells, and time runs vertically down the tree from the root to the tips. (B) Mutational profiles from the scRNA-seq data where 0 denotes the absence of the mutation and 1 denotes the presence of the mutation. The red 0’s (with dashed boxes) denote false negatives, due to either transcriptional bursting or allelic dropout, and the red 1’s (with dashed boxes) denote false positives, due to sequencing errors. (C) The observed and cluster-specific variant allele frequencies (VAFs) associated with the bulk data. (D) When considering only the bulk data, the temporal order of the mutations is inferred correctly, but the branching structure is not. SNV1 and SNV2 have similar observed VAFs, so they co-cluster at the top of the tree producing two structures that differ from the truth by one subclone. (E) When considering only the single-cell data, the branching structure is inferred correctly, but the temporal order of the mutations is incorrect due to noise in the scRNA-seq data. In particular, SNV6 has two false positives that force its placement closer to the root of the tree, and SNV4 has a false negative that forces its placement closer to the tip of the tree

The real strength of Canopy2 lies in its ability to deal with these missing or sparse data points and sort them into categories, such as mutations that are non-cancerous, stochastic (random), or technical issues related to the sequencing process. By doing this, Canopy2 can accurately map the evolutionary tree of tumor cells, even when the data is incomplete or when the tumor is at a very advanced stage with multiple types of mutated cells.

In tests, Canopy2 has outperformed other methods, especially when the data is tricky—such as when there is low sequencing depth or poor cell yield. The tool has been shown to accurately reconstruct the clonal tree (the family tree of tumor cells) in challenging conditions like those seen in breast cancer and glioblastoma. This makes Canopy2 a powerful tool for understanding cancer at a deeper level and potentially discovering more targeted therapies for complex tumors.

Overall, Canopy2 provides researchers with a more reliable and detailed way to study tumor evolution and cell heterogeneity, paving the way for better cancer treatments.

Availability – Canopy2 is an open-source R package available at https://github.com/annweideman/canopy2.