425: BEAM: Bayesian reconstruction of metastatic migration histories

Staklinski SJ et al., Cell Genomics 6, 101193 (2026) - This episode explores BEAM, a Bayesian framework built on BEAST 2 that jointly infers cell-lineage phylogenies and tissue-migration graphs from CRISPR-based lineage-tracing data. The method quantifies uncertainty, improves reconstruction versus parsimony-based approaches, and supports Bayes-factor hypothesis testing of migration models. Applications to simulated data and mouse lung and prostate datasets reveal complex migration patterns and highlight limits imposed by sparse mutational signal. Key terms: Bayesian inference, metastasis, lineage tracing, phylogenetics, BEAM.

Study Highlights:
BEAM jointly samples lineage trees and tissue-migration histories, producing posterior distributions over migration graphs and timing. In simulations BEAM outperforms existing parsimony-based methods across a range of mutation and migration regimes and is robust to missing barcode data. Applied to mouse lung and prostate datasets, BEAM uncovers complex, heterogeneous migration patterns and provides conservative estimates of metastasis-to-metastasis and primary-reseeding events. The framework also implements Bayes-factor tests to assess dataset informativeness and to compare competing migration models.

Conclusion:
BEAM provides a fully Bayesian approach that integrates lineage-tree and migration-graph inference, quantifies uncertainty, and enables formal hypothesis testing; it improves accuracy in many simulated regimes and reveals richer metastatic histories in real datasets, while its utility is constrained by sparse mutational information and current scalability limits.

Music:
Enjoy the music based on this article at the end of the episode.

Article title:
Bayesian inference of tissue-migration histories in metastatic cancer from cell-lineage tracing data

First author:
Staklinski SJ

Journal:
Cell Genomics 6, 101193 (2026)

DOI:
10.1016/j.xgen.2026.101193

Reference:
Staklinski SJ, Scheben A, Brault LM, Hassett R, Serio RN, Xing J, Nowak DG, Siepel A. Bayesian inference of tissue-migration histories in metastatic cancer from cell-lineage tracing data. Cell Genomics. 2026;6:101193. doi:10.1016/j.xgen.2026.101193

License:
This episode is based on an open-access article published under the Creative Commons Attribution 4.0 International License (CC BY 4.0) – https://creativecommons.org/licenses/by/4.0/

Support:
Base by Base is independent and ad-free — no sponsors, no paywall. If an episode was worth your time, chip in and keep the papers audited and the original songs coming:
❤️ Support monthly: https://buy.stripe.com/cNifZhclVebvagk2JDgEg01
☕ One-time donation: https://donate.stripe.com/7sY4gz71B2sN3RWac5gEg00
More at basebybase.com

On PaperCast Base by Base you'll discover the latest in genomics, functional genomics, structural genomics, and proteomics.

Episode link: https://basebybase.com/episodes/beam-bayesian-inference-metastasis

QC:
This episode was checked against the original article PDF and publication metadata for the episode release published on 2026-07-23.

QC Scope:
- article metadata and core scientific claims from the narration
- excludes analogies, intro/outro, and music
- transcript coverage: Audited the transcript's substantive claims about BEAM's methodology, benchmarking against parsimony methods, simulated performance, real-data findings (lung and prostate), Bayes-factor testing, and limitations/future directions; compared against the canonical article text.
- transcript topics: BEAM Bayesian joint inference on BEAST 2; Two-step parsimony methods and their limitations; Simulation benchmarks and edgewise performance; Real-data analyses: lung cancer migration histories; Prostate cancer migration histories and data informativeness; Bayesian hypothesis testing and Bayes fac...

Chapters

• (00:00:02) - Papercast: The Science of Genomics
• (00:00:29) - How does cancer spread? The '
• (00:05:55) - Bayesian Analysis of Cancer metastasis
• (00:10:55) - Beme the Better Way to Map Prostate Cancer?
• (00:15:16) - Beme the computational lung cancer model