Inference of single-cell phylogenies from lineage tracing data using Cassiopeia

Area of Science:

• Computational Biology
• Genetics
• Bioinformatics

Background:

• CRISPR/Cas9 gene editing combined with single-cell readouts allows for large-scale lineage tracing.
• Current data complexity from these assays challenges accurate phylogenetic relationship inference.
• Scalable computational methods are needed to analyze complex lineage tracing datasets.

Purpose of the Study:

• To introduce Cassiopeia, a suite of scalable maximum parsimony methods for phylogenetic tree reconstruction.
• To provide a simulation framework for evaluating lineage tracing algorithms and tracer design.
• To benchmark phylogenetic inference approaches using a large-scale experimental dataset.

Main Methods:

• Development of Cassiopeia, a scalable maximum parsimony-based algorithm for phylogenetic tree reconstruction.
• Creation of a simulation framework for algorithm evaluation and lineage tracer design principles.
• Generation and analysis of a 34,557-cell human lineage tracing dataset over 15 generations.

Main Results:

• Cassiopeia demonstrates superior performance compared to traditional methods across various metrics and parameter settings.
• Benchmarking on a large dataset confirms Cassiopeia's accuracy and scalability.
• Insights into designing improved Cas9-enabled lineage recorders were gained.

Conclusions:

• Cassiopeia provides a robust and scalable solution for inferring phylogenetic relationships from complex lineage tracing data.
• The developed simulation framework aids in algorithm development and experimental design.
• This work facilitates large-scale mammalian lineage tracing efforts and understanding developmental processes.