The Bourque distances for mutation trees of cancers

Area of Science:

• Computational Biology
• Bioinformatics
• Cancer Genomics

Background:

• Mutation trees, also known as clonal trees, model tumor mutational history.
• Traditional tree metrics like Robinson-Foulds distance are inadequate for comparing mutation trees due to differing mutation labels.
• This limitation hinders comparative analyses across different inference methods or patient datasets.

Purpose of the Study:

• To develop novel distance metrics for comparing mutation trees.
• To address the limitations of existing metrics in handling varied mutation labels.
• To facilitate more robust comparisons of tumor evolutionary histories.

Main Methods:

• Generalizing the Robinson-Foulds distance to create a new set of metrics, termed Bourque distances.
• Developing an efficient algorithm for computing the basic Bourque distance.
• Establishing a theoretical link between the Bourque distance and the nearest neighbor interchange distance.

Main Results:

• Introduced Bourque distances, a generalized set of metrics for mutation tree comparison.
• Demonstrated that the basic Bourque distance can be computed in linear time.
• Established a connection between Robinson-Foulds distance and nearest neighbor interchange distance.

Conclusions:

• Bourque distances offer a more effective approach for comparing mutation trees in computational oncology.
• The linear-time computation of the basic Bourque distance enhances its practical applicability.
• These new metrics advance the analysis of tumor evolution and comparative genomics.