Deep distributed computing to reconstruct extremely large lineage trees

Konno, Naoki; Kijima, Yusuke; Watano, Keito; Ishiguro, Shigeo; Ono, Keiichiro; Tanaka, Mamoru; Mori, Hideto; Masuyama, Nanami; Pratt, Dexter; Ideker, Trey; Iwasaki, Wataru; Yachie, Nozomu

doi:10.1038/s41587-021-01111-2

Cited by 19 publications

(11 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, methods were developed that can incorporate a priori information on the frequency of editing outcomes [ 10 – 12 ], which helps reduce biased inference due to homoplasy or the exclusive use of pairwise distances. While some approaches focus on improving scalability to reconstruct trees with millions of cells [ 10 , 13 ], others focus on detailed modelling of the editing process to enable more accurate inference [ 12 ]. A key example of the latter is the maximum-likelihood framework GAPML [ 12 ], which models the editing process of a GESTALT recorder [ 2 ].…”

Section: Introductionmentioning

confidence: 99%

TiDeTree: a Bayesian phylogenetic framework to estimate single-cell trees and population dynamic parameters from genetic lineage tracing data

Seidel

Stadler

2022

Proc. R. Soc. B.

View full text Add to dashboard Cite

The development of organisms and tissues is dictated by an elaborate balance between cell division, apoptosis and differentiation: the cell population dynamics. To quantify these dynamics, we propose a phylodynamic inference approach based on single-cell lineage recorder data. We developed a Bayesian phylogenetic framework—time-scaled developmental trees (TiDeTree)—that uses lineage recorder data to estimate time-scaled single-cell trees. By implementing TiDeTree within BEAST 2, we enable joint inference of the time-scaled trees and the cell population dynamics. We validated TiDeTree using simulations and showed that performance further improves when including multiple independent sources of information into the inference, such as frequencies of editing outcomes or experimental replicates. We benchmarked TiDeTree against state-of-the-art methods and show comparable performance in terms of tree topology, plus direct assessment of uncertainty and co-estimation of additional parameters. To demonstrate TiDeTree’s use in practice, we analysed a public dataset containing lineage data from approximately 100 stem cell colonies. We estimated a time-scaled phylogeny for each colony; as well as the cell division and apoptosis rates underlying the growth dynamics of all colonies. We envision that TiDeTree will find broad application in the analysis of single-cell lineage tracing data, which will improve our understanding of cellular processes during development.

show abstract

Section: Introductionmentioning

confidence: 99%

TiDeTree: a Bayesian phylogenetic framework to estimate single-cell trees and population dynamic parameters from genetic lineage tracing data

Seidel

Stadler

2022

Proc. R. Soc. B.

View full text Add to dashboard Cite

show abstract

“…However, such inferred phylogenetic trees are inherently error-prone for two reasons. First, exhaustive phylogenetic tree search to guarantee optimality is not computationally practical for hundreds of terminal cells, and practical heuristic algorithms do not guarantee optimality [26] despite the recent advances in distributed computing [27]. Second, barcoding strategies employ a limited number of mutation sites which encode a limited amount of information; how close any inferred tree—including the optimal tree—is to its true phylogeny remains uncertain.…”

Section: Resultsmentioning

confidence: 99%

Quantitative fate mapping: Reconstructing progenitor field dynamics via retrospective lineage barcoding

Fang

Bell

Sapirstein

et al. 2022

Preprint

View full text Add to dashboard Cite

Natural and induced somatic mutations that accumulate in the genome during development record the phylogenetic relationships of cells; however, whether these lineage barcodes can capture the dynamics of complex progenitor fields remains unclear. Here, we introduce quantitative fate mapping, an approach to simultaneously map the fate and quantify the commitment time, commitment bias, and population size of multiple progenitor groups during development based on a time-scaled phylogeny of their descendants. To reconstruct time-scaled phylogenies from lineage barcodes, we introduce Phylotime, a scalable maximum likelihood clustering approach based on a generalizable barcoding mutagenesis model. We validate these approaches using realistically-simulated barcoding results as well as experimental results from a barcoding stem cell line. We further establish criteria for the minimum number of cells that must be analyzed for robust quantitative fate mapping. Overall, this work demonstrates how lineage barcodes, natural or synthetic, can be used to obtain quantitative fate maps, thus enabling analysis of progenitor dynamics long after embryonic development in any organism.

show abstract

“…This includes genomic datasets with many samples from an individual pathogen, including for example large collections of M. tuberculosis genomes 43 or influenza genomes 44 , and collections of genomic data from possible future pandemics. Our approach could also be combined with divide-and-conquer phylogenetic algorithms 45,46 to further improve its performance and applicability.…”

Section: Discussionmentioning

confidence: 99%

Maximum likelihood pandemic-scale phylogenetics

Maio

Kalaghatgi

Turakhia

et al. 2022

Preprint

View full text Add to dashboard Cite

Genomic data plays an essential role in the study of transmissible disease, as exemplified by its current use in identifying and tracking the spread of novel SARS-CoV-2 variants. However, with the increase in size of genomic epidemiological datasets, their phylogenetic analyses become increasingly impractical due to high computational demand. In particular, while maximum likelihood methods are go-to tools for phylogenetic inference, the scale of datasets from the ongoing pandemic has made apparent the urgent need for more computationally efficient approaches. Here we propose a new likelihood-based phylogenetic framework that greatly reduces both the memory and time demand of popular maximum likelihood approaches when analysing many closely related genomes, as in the scenario of SARS-CoV-2 genome data and more generally throughout genomic epidemiology. To achieve this, we rewrite the classical Felsenstein pruning algorithm so that we can infer phylogenetic trees on at least 10 times larger datasets with higher accuracy than existing maximum likelihood methods. Our algorithms provide a powerful framework for maximum-likelihood genomic epidemiology and could facilitate similarly groundbreaking applications in Bayesian phylogenomic analyses as well.

show abstract

Deep distributed computing to reconstruct extremely large lineage trees

Cited by 19 publications

References 50 publications

TiDeTree: a Bayesian phylogenetic framework to estimate single-cell trees and population dynamic parameters from genetic lineage tracing data

TiDeTree: a Bayesian phylogenetic framework to estimate single-cell trees and population dynamic parameters from genetic lineage tracing data

Quantitative fate mapping: Reconstructing progenitor field dynamics via retrospective lineage barcoding

Maximum likelihood pandemic-scale phylogenetics

Contact Info

Product

Resources

About