SciPhy: A Bayesian phylogenetic framework using sequential genetic lineage tracing data
12 affiliations
- 1Department of Biosystems Science and Engineering, ETH Zürich, Basel, Switzerland. [email protected]
- 2Swiss Institute of Bioinformatics, Lausanne, Switzerland. [email protected]
- 3Department of Biosystems Science and Engineering, ETH Zürich, Basel, Switzerland
- 4Swiss Institute of Bioinformatics, Lausanne, Switzerland
- 5Department of Genome Sciences, University of Washington, Seattle, WA, USA
- 6Developmental Biology Program, Memorial Sloan Kettering Cancer Center, New York, NY, USA
- 7Allen Discovery Center for Cell Lineage Tracing, Seattle, WA, USA
- 8Brotman Bay Institute for Precision Medicine, Seattle, WA, USA
- 9Howard Hughes Medical Institute, Seattle, WA, USA
- 10Seattle Hub for Synthetic Biology, Seattle, WA, USA
- 11Department of Biosystems Science and Engineering, ETH Zürich, Basel, Switzerland. [email protected]
- 12Swiss Institute of Bioinformatics, Lausanne, Switzerland. [email protected]
Abstract
CRISPR-based lineage tracing offers a promising avenue to decipher single cell lineage trees, especially in organisms that are challenging for microscopy. A recent advancement in this domain is lineage tracing based on sequential genome editing, which not only records genetic edits but also the order in which they occur. To capitalize on this enriched data, we introduce SciPhy, a simulation and inference tool integrated within the BEAST 2 framework. SciPhy utilizes a Bayesian phylogenetic approach to estimate time-scaled phylogenies and cell population parameters. After validating SciPhy using simulations, we apply it to two lineage tracing datasets for which we estimate time-scaled trees together with cell proliferation rates. Using simulated and real lineage tracing data obtained from a monoclonal culture of HEK293T cells, we compare SciPhy to other lineage reconstruction methods, and find that SciPhy consistently constructs distinct, and more accurate lineage trees. In particular, for HEK293T cells, SciPhy trees stand out for their later estimated cell division times. In addition, SciPhy reports uncertainty as well as proliferation rates, neither of which are available within a UPGMA analysis. Our second example applies SciPhy to the study of murine gastruloids, and showcases the use of complex models of time-varying population growth to capture realistic aspects of this in-vitro model of early mammalian development. Together, these examples showcase the application of advanced phylogenetic and phylodynamic tools to explore and quantify cell lineage trees, laying the groundwork for enhanced and confident analyses to decode the complexities of biological development in multicellular organisms. SciPhy’s codebase is publicly available at https://github.com/azwaans/SciPhy .
Related
- Evolving strategies for lineage tracing: Genetic markers, synthetic barcodes, and natural variants
- PhyloZoo: a unified framework for phylogenetic network analysis in Python.
- Chroma-Agent-Alpha: An Open-Source Framework for Automated Chromatography Preprocessing, GNN-Based Peak Deconvolution, and FAIR Data Lineage
- A multi-objective sequence engineering framework for rational genetic vaccine design.
- A Bayesian framework for multivariate differential analysis
BibTeX
@article{seidel2026sciphy,
title = {{SciPhy: A Bayesian phylogenetic framework using sequential genetic lineage tracing data}},
author = {Sophie Seidel and Antoine Zwaans and Samuel G. Regalado and Junhong Choi and Jay Shendure and Tanja Stadler},
journal = {Nature Communications},
year = {2026},
month = jun,
doi = {10.1038/s41467-026-73377-6},
url = {https://doi.org/10.1038/s41467-026-73377-6},
note = {CC BY},
}
RIS
TY - JOUR
TI - SciPhy: A Bayesian phylogenetic framework using sequential genetic lineage tracing data
AU - Sophie Seidel
AU - Antoine Zwaans
AU - Samuel G. Regalado
AU - Junhong Choi
AU - Jay Shendure
AU - Tanja Stadler
PY - 2026
DA - 2026/06/10
JO - Nature Communications
DO - 10.1038/s41467-026-73377-6
UR - https://doi.org/10.1038/s41467-026-73377-6
AB - CRISPR-based lineage tracing offers a promising avenue to decipher single cell lineage trees, especially in organisms that are challenging for microscopy. A recent advancement in this domain is lineage tracing based on sequential genome editing, which not only records genetic edits but also the order in which they occur. To capitalize on this enriched data, we introduce SciPhy, a simulation and inference tool integrated within the BEAST 2 framework. SciPhy utilizes a Bayesian phylogenetic approach to estimate time-scaled phylogenies and cell population parameters. After validating SciPhy using simulations, we apply it to two lineage tracing datasets for which we estimate time-scaled trees together with cell proliferation rates. Using simulated and real lineage tracing data obtained from a monoclonal culture of HEK293T cells, we compare SciPhy to other lineage reconstruction methods, and find that SciPhy consistently constructs distinct, and more accurate lineage trees. In particular, for HEK293T cells, SciPhy trees stand out for their later estimated cell division times. In addition, SciPhy reports uncertainty as well as proliferation rates, neither of which are available within a UPGMA analysis. Our second example applies SciPhy to the study of murine gastruloids, and showcases the use of complex models of time-varying population growth to capture realistic aspects of this in-vitro model of early mammalian development. Together, these examples showcase the application of advanced phylogenetic and phylodynamic tools to explore and quantify cell lineage trees, laying the groundwork for enhanced and confident analyses to decode the complexities of biological development in multicellular organisms. SciPhy’s codebase is publicly available at https://github.com/azwaans/SciPhy .
SN - 2041-1723
ER -