Comparing Phylogenetic Approaches to Reconstructing Cell Lineage From Microsatellites With Missing Data.

Area of Science:

• Genomics
• Computational Biology
• Evolutionary Biology

Background:

• Somatic cells accumulate DNA mutations during cell division, recording their lineage history.
• Microsatellites, highly mutable DNA regions, are key markers for tracing cell lineages.
• Advances in single-cell sequencing and phylogenetic methods enable detailed lineage reconstruction.

Purpose of the Study:

• To evaluate computational approaches for reconstructing cell lineage trees from single-cell Next Generation Sequencing (NGS) data.
• To identify the most accurate and robust phylogenetic algorithms for this purpose.

Main Methods:

• Simulated cell division trees with mutating microsatellites.
• Tested various phylogenetic algorithms, including distance-based, Maximum Parsimony, and probabilistic methods.
• Assessed algorithm robustness against different tree topologies and mutation models.

Main Results:

• Distance-based methods are fast and accurate with complete data.
• Maximum Parsimony and probabilistic methods show greater robustness with missing data, making them suitable for NGS datasets.
• Reconstruction accuracy is adaptable across various biological scenarios.

Conclusions:

• Maximum Parsimony and probabilistic phylogenetic approaches are recommended for cell lineage reconstruction from single-cell NGS data.
• These methods offer flexibility and robustness, crucial for handling real-world biological data complexities.