Pith. sign in

REVIEW 3 major objections 4 minor 1 cited by

Likelihood-free inference of phylogenetic tree posterior distributions

T0 review · 3 major / 4 minor · reviewed 2026-08-04 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Phyloformer 2 is the first likelihood-free method to estimate full posterior distributions over phylogenetic trees directly from aligned sequences, and it reconstructs topologies more accurately than likelihood-based state-of-the-art method

desk verdict Genuinely new likelihood-free posterior estimator for full phylogenies, but the missing final edge in the merge factorization leaves the normalization claim unproven — fixable, but central. read the letter →

arxiv 2510.12976 v4 pith:72WUPUR5 submitted 2025-10-14 q-bio.PE q-bio.QM

Luc Blassel , Noémie Sauvage , Pierre Barrat-Charlaix , Bastien Boussau , Nicolas Lartillot , Laurent Jacob This is my paper · ORCID
classification q-bio.PEq-bio.QM
keywords phylogeneticinferencelikelihood-freeneuralposteriorestimationdeeplearningBayesianphylogeneticssequenceevolutionamortizedBayesNJ
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper introduces Phyloformer 2, a neural network that estimates the entire posterior distribution over phylogenetic trees without ever computing a likelihood. It trains on simulated alignments and represents the posterior as a product of conditional probabilities over a sequence of subtree merges, with branch lengths modeled by parametric distributions. On tractable sequence-evolution models the method gives more accurate tree topologies than leading maximum-likelihood tools, provides well-calibrated branch-support estimates, and runs one to two orders of magnitude faster at inference time. Its main payoff is that it can be trained under evolutionary models with intractable likelihoods, where its accuracy advantage over misspecified likelihood-based methods grows.

What carries the argument

BayesNJ: a parameterized probability distribution over phylogenies factorized over successive pairwise merges, using a softmin over symmetric bilinear pair scores for the topological choice and shifted Gamma and Beta distributions for branch lengths, with a constraint mechanism enforcing a canonical merge order so the product is a proper distribution over trees. evoPF: an encoder maintaining one embedding per sequence and one per pair of sequences, sharing information through self-attention and outer-product means, which scales to hundreds of sequences.

What would settle it

Train Phyloformer 2 under a birth-death prior and LG+G8, then apply it to alignments simulated under a different prior (e.g., uniform topology with exponential branch lengths) or to real empirical alignments. If topological accuracy degrades sharply and calibration curves move off the diagonal while a likelihood-based method with model selection degrades gracefully, the claimed edge is an artifact of prior matching rather than a general property of the estimator.

Watch

Extended reading notes

Core claim

The authors claim that a neural posterior estimator can replace likelihood computation for Bayesian phylogenetic inference. Phyloformer 2 combines evoPF, an EvoFormer-style encoder that turns aligned sequences into per-sequence and per-pair embeddings, with BayesNJ, a factorization of the tree posterior over a canonical sequence of merges. Under a tractable model (LG+G8), its greedy MAP trees have lower normalized Robinson-Foulds distance to the true tree than maximum-likelihood methods; under models with intractable likelihoods (Cherry and SelReg), the gap widens. Comparison to a long MCMC run on a 50-taxon alignment shows strong agreement in split frequencies, and simulation-based calibrat

Load-bearing premise

The whole pipeline assumes that the data were generated by the same prior over trees and the same sequence-evolution model used to simulate training data; the authors state that outside that distribution the network would currently produce poor estimates with no warning.

Editorial extensions

If this is right

  • Once trained, Phyloformer 2 samples from the posterior orders of magnitude faster than MCMC, making full posterior analysis practical on large datasets.
  • Because it is likelihood-free, it can be trained under richer evolutionary models—with site dependencies, heterogeneous selection, or other complex generative processes—where likelihood computation is infeasible.
  • Branch posterior probabilities from the network are well calibrated (slightly conservative), so they can be used as uncertainty estimates for reconstructed clades.
  • When the training prior matches the data-generating process, the method outperforms maximum-likelihood tools even under tractable models, and its edge grows when the likelihood model is misspecified.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The claimed performance is conditional on the training prior; the authors themselves note that inputs far from training data would produce poor estimates with no warning, so practical deployment would need an out-of-distribution detector or uncertainty layer.
  • The canonical-merge trick could generalize to other recursive combinatorial objects in biology, such as gene-tree/species-tree reconciliations or phylogenetic networks, wherever a canonical decomposition can be enforced.
  • Posterior quality is currently evaluated mainly through split frequencies; a stricter test would compare full tree distributions or clade probabilities on larger alignments, which could reveal whether the merge-wise factorization loses higher-order correlations.
  • Because the ablation suggests most of the topological gain comes from the BayesNJ loss rather than the encoder alone, making the branch-length distributions more flexible than Gamma/Beta could further improve calibration.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 4 minor

Summary. The paper presents Phyloformer 2, a neural posterior estimator for phylogenies from aligned sequence data. It combines an evoPF encoder (an EvoFormer-like architecture) with a BayesNJ module that defines a factorized distribution over trees through a succession of pairwise merges. The authors claim this is the first likelihood-free posterior estimation method trained end-to-end beyond quartets, and report improved topological accuracy over likelihood-based tools, faster inference, and well-calibrated posterior branch supports on simulated data. The architecture, training procedure, and experiments are described in detail, with supplementary algorithms and extensive benchmarking.

Significance. If the mathematical inconsistencies in the definition of BayesNJ are resolved, this would be a substantial contribution: it demonstrates a scalable amortized likelihood-free approach to full phylogenetic posterior inference, with promising empirical performance and a clear path to use under intractable-likelihood models. The paper includes extensive experiments and comparisons, and the qualitative empirical claims are supported by the data in the relevant regimes. The issues identified here concern the formal correctness of the model distribution and are essential to address before the posterior calibration claims can be taken at face value.

major comments (3)
  1. [§3.2, Eq. (1) and Algorithms S.10/S.11] The product defining qψ is over 2N−3 merges, but Algorithm S.10 and S.11 loop only over k=1..N−2 and emit two branch lengths per merge, i.e., 2N−4 branch lengths. The final edge connecting the two remaining active nodes is never assigned a length, so qψ does not define a probability density on R_+^{2N−3}. If the final edge is intended to be handled implicitly, it must be stated explicitly and included in the density; otherwise the parameter space of the model is not the space of phylogenies.
  2. [§3.2 and Algorithm S.11] The constraint procedure does not guarantee that the sampling order matches the canonical merge order used for evaluation. The constraints only impose that each sampled cherry sum be at least the sum of any previous merge performed while that pair was available; they do not ensure that the sampled pair is the minimum-sum pair among all current candidates in the final tree. Concretely, for N=4, a tree with s(a,b)=20 and a later cherry (u1,c) with sum 12 would be sampled in the order (a,b) first, whereas the canonical order of the same tree would merge (a,c) first. Hence the sampled order and the evaluation order can disagree, and the same tree can receive different probabilities or be missed entirely. This breaks the normalization argument for qψ.
  3. [§3.3, Algorithm S.10, Appendix A.2] The reparameterization of the branch-length distribution is inconsistent between the main text, the training loss, and the appendix. §3.3 defines r = ℓ_i/(ℓ_i+ℓ_j) and states that the Jacobian factor is 1/s. Algorithm S.10 evaluates the Beta density at ℓ_i/ℓ_j (not at r) and omits the Jacobian factor entirely. Algorithm S.11 samples r from the Beta and forms ℓ_i = r·s, ℓ_j = s−ℓ_i, which corresponds to the main-text reparameterization. Appendix A.2 gives the determinant as −1/r (should be −1/s) and concludes qℓ = q_s q_r / r. These three descriptions cannot all be correct, and without code it is impossible to determine which objective was actually optimized. This ambiguity affects the branch-length posterior and, through the training objective, the entire posterior approximation.
minor comments (4)
  1. [Abstract] The claim of 'more accurate tree topologies than existing methods' is unqualified; Figure 2a shows this holds for 10–175 leaves, but at 200 leaves the advantage over IQTree is not demonstrated. Please qualify the abstract accordingly.
  2. [§4.1] The test set is simulated under LG+GC while PF2 is trained on LG+G8. This is a model mismatch that should be explicitly acknowledged in the main text and accounted for when interpreting the comparison.
  3. [Algorithm S.11] The line 'ℓ(k)j ← s(k) − ℓ(k)j' appears to be a typo: it should read 'ℓ(k)j ← s(k) − ℓ(k)i'.
  4. [Code availability] The paper does not state code availability; given the discrepancies in the algorithms, releasing the implementation would greatly facilitate verification.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: the posterior estimator is benchmarked against external likelihood-based and MCMC tools; minor self-citations provide priors/data, not the load-bearing argument.

full rationale

The central claims of the paper are empirical and are evaluated against external benchmarks: topological accuracy is compared with IQTree, FastTree, FastME, and Phyloformer; posterior distributions are compared with RevBayes MCMC; calibration is checked against held-out simulated trees. These tests do not reduce to the model's own training objective by construction. The NPE training procedure maximizes qψ on (x, θ) pairs, and the fact that test data come from the same generative model is a standard amortized-inference setup, not a fitted parameter renamed as a prediction. The paper explicitly acknowledges that its advantage over likelihood-based methods likely arises from using the correct prior, which is an assumption about the evaluation setting rather than a circular step. Self-citations to Nesterenko et al. (2025) supply simulation priors, datasets, and baseline numbers, but they do not justify the main inference claims, so they are not load-bearing. The internal concern about the BayesNJ factorization—the product over 2N−3 merges while the algorithms loop N−2 times and emit 2N−4 branch lengths, plus the Jacobian inconsistency in Appendix A.2— is a correctness and verifiability issue about whether qψ is a normalized density, not a circular reduction of the kind defined here. Overall, no equation is shown to be equivalent to its inputs by construction, and the paper is self-contained against external likelihood-based tools.

Assumptions & free parameters 3 free parameters · 6 assumptions · 0 invented entities

The method introduces no new physical or conceptual entities; the learned parent-node embeddings and the BayesNJ factorized distribution are functions of the input data and model parameters, not independent postulated entities. The main burden is the ad hoc Gamma/Beta branch-length families and the canonical merge-order construction, plus the standard NPE training objective and the domain assumption that test data resemble the training distribution.

free parameters (3)
  • evoPF/BayesNJ network weights = trained on ~7.9M 50-taxa LG+G8 MSAs + fine-tuning
    The posterior distribution parameters are all outputs of learned neural networks; this is an NPE method, so the 'posterior' is a learned fit, not a derivation.
  • Number of evoPF blocks = 12
    Chosen by hand; no ablation study justifies this value.
  • Embedding dimensions (cs, cz) = 128, 256
    Chosen by hand; no sensitivity analysis is reported.
assumptions (6)
  • standard math NPE training objective: minimizing average KL between q_psi and p(theta|x) is equivalent to maximizing E_{p(x,theta)}[log q_psi(theta|x)] (Section 2.2).
    The paper relies on the standard NPE convergence result; no formal proof is given, but it is textbook.
  • ad hoc to paper The canonical merge order (at each step merge the pair with smallest sum of branch lengths) defines a unique merge sequence per tree, and the constraint mechanism makes the factorized distribution proper.
    This is the paper's key modeling innovation; it is asserted (Section 3.2) and not proven rigorously, and the evaluation procedure for an arbitrary tree is underspecified.
  • ad hoc to paper Branch-length posteriors are modeled as a shifted Gamma (sum) and a Beta (ratio).
    The paper states in Section 3.3 that 'the true posterior has no reason to match these analytical forms in general,' an acknowledged limitation.
  • domain assumption Training and test data are generated under similar (but not identical) priors; the method's accuracy and calibration presuppose the test distribution is close to the training distribution.
    This is the weakest load-bearing assumption: Section 5 admits out-of-distribution inputs produce poor estimates with no warning; the main test set (LG+GC) differs from training (LG+G8).
  • domain assumption The Cherry benchmark, despite its known simulation error, still serves as a proof-of-concept for intractable-likelihood inference.
    Appendix A.5 documents the Cherry simulation flaw; the authors choose to keep it in the evaluation.
  • domain assumption Input sequences are aligned; the method does not handle unaligned sequences.
    Section 2.1 assumes aligned sequences; unaligned inference is listed as future work.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Likelihood-free inference of phylogenetic tree posterior distributions." pith.science (2026). https://pith.science/paper/72WUPUR5

@misc{pith2026251012976,
  author       = {Pith},
  title        = {Pith review of: Likelihood-free inference of phylogenetic tree posterior distributions},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/72WUPUR5}},
  note         = {Machine review of arXiv:2510.12976}
}
read the original abstract

Phylogenetic inference, the task of reconstructing how related sequences evolved from common ancestors, is a central objective in evolutionary genomics. The current state-of-the-art methods exploit probabilistic models of sequence evolution along phylogenetic trees, by searching for the tree maximizing the likelihood of observed sequences, or by estimating the posterior of the tree given the sequences in a Bayesian framework. Both approaches typically require to compute likelihoods, which is only feasible under simplifying assumptions such as independence of the evolution at the different positions of the sequence, and even then remains a costly operation. Here we present the first likelihood-free inference method for posterior distributions over phylogenies. It exploits a novel expressive encoding for pairs of sequences, and a parameterized probability distribution factorized over a succession of subtree merges. The resulting network provides well-calibrated estimates of the posterior distribution leading to more accurate tree topologies than existing methods, even under models amenable to likelihood computation. We further show that its edge against likelihood-based methods dramatically increases under models of sequence evolution with intractable likelihoods.

Figures

Figures reproduced from arXiv: 2510.12976 by the authors.

Figure 1
Figure 1. Architecture of Phyloformer 2. Panel a: evoPF, an EvoFormer￾inspired module updating N ×L embeddings for a set of aligned sequences (MSA) x and N(N − 1)/2 embeddings for pairs of sequences. Each of its blocks applies self-attention within both the MSA and representation, and ensures information sharing between them. After 12 blocks, we extract one embedding for each sequence by averaging the MSA embeddings across si… view at source ↗
Figure 2
Figure 2. (a) Topological performance for Phyloformer 2, measured by the normalized Robinson-Foulds distance. The alignments for which trees were inferred were taken from the original Phyloformer paper (Nesterenko et al., 2025) and were simulated under the LG+GC sequence model. (b) Runtime and (c) Memory usage for Phyloformer 2. The same GPU model as the original Phyloformer study was used to run Phyloformer 2 inference. Resu… view at source ↗
Figure 3
Figure 3. a): Comparison of split frequencies over samples from the posterior of a single 50 sequences MSA, between RevBayes MCMC (x-axis) and PF2 (y-axis). The orange cross marker indicates splits that are present in the tree along which the MSA used for sampling has been simulated. b): simulation-based calibration comparing branch supports, i.e., their frequency in samples from the posterior estimate given by a trained PF2 … view at source ↗

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Modeling Protein Evolution with Generative Models: from Extant Sequence Data to Evolutionary Dynamics

    q-bio.PE 2026-06 unverdicted

    Review of generative sequence models and Direct Coupling Analysis for simulating protein evolutionary dynamics from extant data.

Reference graph

Works this paper leans on

44 extracted references · 10 canonical work pages · cited by 1 Pith paper

  1. [1]

    Aminov and Roderick I

    Rustam I. Aminov and Roderick I. Mackie. Evolution and ecology of antibiotic resistance genes. FEMS Microbiology Letters, 271 0 (2): 0 147--161, 06 2007. ISSN 0378-1097. doi:10.1111/j.1574-6968.2007.00757.x. URL https://doi.org/10.1111/j.1574-6968.2007.00757.x

    arXiv 2007

  2. [2]

    Alexandre Bouchard-Côté, Sriram Sankararaman, and Michael I. Jordan. Phylogenetic inference via sequential monte carlo. Systematic Biology, 61 0 (4): 0 579--593, 01 2012. ISSN 1063-5157. doi:10.1093/sysbio/syr131. URL https://doi.org/10.1093/sysbio/syr131

    work page doi:10.1093/sysbio/syr131 2012

  3. [3]

    The frontier of simulation-based inference

    Kyle Cranmer, Johann Brehmer, and Gilles Louppe. The frontier of simulation-based inference. Proceedings of the National Academy of Sciences, 117 0 (48): 0 30055--30062, 2020. doi:10.1073/pnas.1912789117. URL https://www.pnas.org/doi/abs/10.1073/pnas.1912789117

    doi:10.1073/pnas.1912789117 2020

  4. [4]

    ChenRui Duan, Zelin Zang, Siyuan Li, Yongjie Xu, and Stan Z. Li. Phylogen: Language model-enhanced phylogenetic inference via graph structure generation. In The Thirty-eighth Annual Conference on Neural Information Processing Systems, 2024. URL https://openreview.net/forum?id=GxvDsFArxY

    2024

  5. [5]

    Evaluation of methods to detect shifts in directional selection at the genome scale

    Louis Duchemin, Vincent Lanore, Philippe Veber, and Bastien Boussau. Evaluation of methods to detect shifts in directional selection at the genome scale. Molecular Biology and Evolution, 40 0 (2): 0 msac247, 12 2022. ISSN 1537-1719. doi:10.1093/molbev/msac247. URL https://doi.org/10.1093/molbev/msac247

    work page doi:10.1093/molbev/msac247 2022

  6. [6]

    Evolutionary trees from dna sequences: a maximum likelihood approach

    Joseph Felsenstein. Evolutionary trees from dna sequences: a maximum likelihood approach. Journal of molecular evolution, 17 0 (6): 0 368--376, 1981

    1981

  7. [7]

    Dropout as a bayesian approximation: Representing model uncertainty in deep learning

    Yarin Gal and Zoubin Ghahramani. Dropout as a bayesian approximation: Representing model uncertainty in deep learning. In Maria Florina Balcan and Kilian Q. Weinberger (eds.), Proceedings of The 33rd International Conference on Machine Learning, volume 48 of Proceedings of Machine Learning Research, pp.\ 1050--1059, New York, New York, USA, 20--22 Jun 201...

    2016

  8. [8]

    Automatic posterior transformation for likelihood-free inference

    David Greenberg, Marcel Nonnenmacher, and Jakob Macke. Automatic posterior transformation for likelihood-free inference. In Kamalika Chaudhuri and Ruslan Salakhutdinov (eds.), Proceedings of the 36th International Conference on Machine Learning, volume 97 of Proceedings of Machine Learning Research, pp.\ 2404--2414. PMLR, 09--15 Jun 2019. URL https://proc...

    2019

Show all 44 references
  1. [9]

    Grosshauser M, Zaharias P

    Warnow T. Grosshauser M, Zaharias P. Re-evaluating deep neural networks for phylogeny estimation: the issue of taxon sampling. Recomb2021, 2021

    2021

  2. [10]

    Nextstrain: real-time tracking of pathogen evolution

    James Hadfield, Colin Megill, Sidney M Bell, John Huddleston, Barney Potter, Charlton Callender, Pavel Sagulenko, Trevor Bedford, and Richard A Neher. Nextstrain: real-time tracking of pathogen evolution. Bioinformatics, 34 0 (23): 0 4121--4123, December 2018. ISSN 1367-4803. ...

    2018 doi

  3. [11]

    Axial attention in multidimensional transformers

    Jonathan Ho, Nal Kalchbrenner, Dirk Weissenborn, and Tim Salimans. Axial attention in multidimensional transformers. CoRR, abs/1912.12180, 2019. URL http://arxiv.org/abs/1912.12180

    1912 arXiv

  4. [12]

    Huelsenbeck and Fredrik Ronquist

    John P. Huelsenbeck and Fredrik Ronquist. Mrbayes: Bayesian inference of phylogenetic trees. Bioinformatics, 17 0 (8): 0 754--755, 08 2001. ISSN 1367-4803. doi:10.1093/bioinformatics/17.8.754. URL https://doi.org/10.1093/bioinformatics/17.8.754

    2001 doi

  5. [13]

    Landis, Tracy A

    Sebastian Höhna, Michael J. Landis, Tracy A. Heath, Bastien Boussau, Nicolas Lartillot, Brian R. Moore, John P. Huelsenbeck, and Fredrik Ronquist. Revbayes: Bayesian phylogenetic inference using graphical models and an interactive model-specification language. Systematic Biolo...

    2016 doi

  6. [14]

    Scaling DEPP phylogenetic placement to ultra-large reference trees: a tree-aware ensemble approach

    Yueyu Jiang, Daniel McDonald, Daniela Perry, Rob Knight, and Siavash Mirarab. Scaling DEPP phylogenetic placement to ultra-large reference trees: a tree-aware ensemble approach. Bioinformatics, 40 0 (6), June 2024

    2024

  7. [15]

    John Jumper, Richard Evans, Alexander Pritzel, Tim Green, Michael Figurnov, Olaf Ronneberger, Kathryn Tunyasuvunakool, Russ Bates, Augustin Z \'i dek, Anna Potapenko, Alex Bridgland, Clemens Meyer, Simon A. A. Kohl, Andrew J. Ballard, Andrew Cowie, Bernardino Romera-Paredes, S...

    2021

  8. [16]

    Paschalia Kapli, Ziheng Yang, and Maximilian J. Telford. Phylogenetic tree building in the genomic age. Nature Reviews Genetics, 21 0 (7): 0 428--444, July 2020. ISSN 1471-0064. doi:10.1038/s41576-020-0233-0. URL http://www.nature.com/articles/s41576-020-0233-0. Publisher: Nat...

    2020 doi

  9. [17]

    Transformers are rnns: Fast autoregressive transformers with linear attention, 2020

    Angelos Katharopoulos, Apoorv Vyas, Nikolaos Pappas, and Fran c ois Fleuret. Transformers are rnns: Fast autoregressive transformers with linear attention, 2020

    2020

  10. [18]

    Simple and scalable predictive uncertainty estimation using deep ensembles

    Balaji Lakshminarayanan, Alexander Pritzel, and Charles Blundell. Simple and scalable predictive uncertainty estimation using deep ensembles. In I. Guyon, U. Von Luxburg, S. Bengio, H. Wallach, R. Fergus, S. Vishwanathan, and R. Garnett (eds.), Advances in Neural Information P...

    2017

  11. [19]

    A Bayesian mixture model for across-site heterogeneities in the amino-acid replacement process

    Nicolas Lartillot and Hervé Philippe. A Bayesian mixture model for across-site heterogeneities in the amino-acid replacement process. Molecular Biology and Evolution, 21 0 (6): 0 1095--1109, June 2004. ISSN 0737-4038. doi:10.1093/molbev/msh112

    2004 doi

  12. [20]

    Inferring long-term effective population size with mutation–selection models

    Thibault Latrille, Vincent Lanore, and Nicolas Lartillot. Inferring long-term effective population size with mutation–selection models. Molecular Biology and Evolution, 38 0 (10): 0 4573--4587, 06 2021. ISSN 1537-1719. doi:10.1093/molbev/msab160. URL https://doi.org/10.1093/mo...

    2021 doi

  13. [21]

    An Improved General Amino Acid Replacement Matrix

    Si Quang Le and Olivier Gascuel. An Improved General Amino Acid Replacement Matrix . Molecular Biology and Evolution, 25 0 (7): 0 1307--1320, July 2008. ISSN 0737-4038. doi:10.1093/molbev/msn067. URL https://doi.org/10.1093/molbev/msn067

    2008 doi

  14. [22]

    Fastme 2.0: a comprehensive, accurate, and fast distance-based phylogeny inference program

    Vincent Lefort, Richard Desper, and Olivier Gascuel. Fastme 2.0: a comprehensive, accurate, and fast distance-based phylogeny inference program. Molecular biology and evolution, 32 0 (10): 0 2798--2800, 2015

    2015

  15. [23]

    Benchmarking simulation-based inference

    Jan-Matthis Lueckmann, Jan Boelts, David Greenberg, Pedro Goncalves, and Jakob Macke. Benchmarking simulation-based inference. In Arindam Banerjee and Kenji Fukumizu (eds.), Proceedings of The 24th International Conference on Artificial Intelligence and Statistics, volume 130 ...

    2021

  16. [24]

    Iq-tree 2: New models and efficient methods for phylogenetic inference in the genomic era

    Bui Quang Minh, Heiko A Schmidt, Olga Chernomor, Dominik Schrempf, Michael D Woodhams, Arndt von Haeseler, and Robert Lanfear. Iq-tree 2: New models and efficient methods for phylogenetic inference in the genomic era. Molecular Biology and Evolution, 37 0 (5): 0 1530--1534, 02...

    2020 doi

  17. [25]

    Muñoz-Gómez, Edward Susko, Kelsey Williamson, Laura Eme, Claudio H

    Sergio A. Muñoz-Gómez, Edward Susko, Kelsey Williamson, Laura Eme, Claudio H. Slamovits, David Moreira, Purificación López-García, and Andrew J. Roger. Site-and-branch-heterogeneous analyses of an expanded dataset favour mitochondria as sister to known Alphaproteobacteria . Na...

    2022 doi

  18. [26]

    Phyloformer: Fast, accurate, and versatile phylogenetic reconstruction with deep neural networks

    Luca Nesterenko, Luc Blassel, Philippe Veber, Bastien Boussau, and Laurent Jacob. Phyloformer: Fast, accurate, and versatile phylogenetic reconstruction with deep neural networks. Molecular Biology and Evolution, 42 0 (4): 0 msaf051, 03 2025. ISSN 1537-1719. doi:10.1093/molbev...

    2025 doi

  19. [27]

    Price, Paramvir S

    Morgan N. Price, Paramvir S. Dehal, and Adam P. Arkin. Fasttree 2 – approximately maximum-likelihood trees for large alignments. PLOS ONE, 5 0 (3): 0 1--10, 03 2010. doi:10.1371/journal.pone.0009490. URL https://doi.org/10.1371/journal.pone.0009490

    2010 doi

  20. [28]

    Sebastian Prillo, Yun Deng, Pierre Boyeau, Xingyu Li, Po-Yen Chen, and Yun S. Song. Cherryml: scalable maximum likelihood estimation of phylogenetic models. Nature Methods, 20 0 (8): 0 1232--1236, Aug 2023. ISSN 1548-7105. doi:10.1038/s41592-023-01917-9. URL https://doi.org/10...

    2023 doi

  21. [29]

    Radev, Ulf K

    Stefan T. Radev, Ulf K. Mertens, Andreas Voss, Lynton Ardizzone, and Ullrich K \"o the. BayesFlow : Learning complex stochastic models with invertible neural networks. IEEE transactions on neural networks and learning systems, 33 0 (4): 0 1452--1466, 2020

    2020

  22. [30]

    Msa transformer

    Roshan M Rao, Jason Liu, Robert Verkuil, Joshua Meier, John Canny, Pieter Abbeel, Tom Sercu, and Alexander Rives. Msa transformer. In Marina Meila and Tong Zhang (eds.), Proceedings of the 38th International Conference on Machine Learning, volume 139 of Proceedings of Machine ...

    2021

  23. [31]

    Robinson and L.R

    D.F. Robinson and L.R. Foulds. Comparison of phylogenetic trees. Mathematical Biosciences, 53 0 (1): 0 131--147, 1981. ISSN 0025-5564. doi:https://doi.org/10.1016/0025-5564(81)90043-2. URL https://www.sciencedirect.com/science/article/pii/0025556481900432

    1981

  24. [32]

    The neighbor-joining method: a new method for reconstructing phylogenetic trees

    Naruya Saitou and Masatoshi Nei. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Molecular biology and evolution, 4 0 (4): 0 406--425, 1987

    1987

  25. [33]

    Accurate Inference of Tree Topologies from Multiple Sequence Alignments Using Deep Learning

    Anton Suvorov, Joshua Hochuli, and Daniel R Schrider. Accurate Inference of Tree Topologies from Multiple Sequence Alignments Using Deep Learning . Systematic Biology, 69 0 (2): 0 221--233, 09 2019. ISSN 1063-5157. doi:10.1093/sysbio/syz060. URL https://doi.org/10.1093/sysbio/syz060

    2019 doi

  26. [34]

    Novel symmetry-preserving neural network model for phylogenetic inference

    Xudong Tang, Leonardo Zepeda-Nuñez, Shengwen Yang, Zelin Zhao, and Claudia Solís-Lemus. Novel symmetry-preserving neural network model for phylogenetic inference . Bioinformatics Advances, 4 0 (1): 0 vbae022, 02 2024. ISSN 2635-0041. doi:10.1093/bioadv/vbae022. URL https://doi...

    2024 doi

  27. [35]

    Consideration of RNA Secondary Structure Significantly Improves Likelihood - Based Estimates of Phylogeny : Examples from the Bilateria

    Maximilian J Telford, Michael J Wise, and Vivek Gowri-Shankar. Consideration of RNA Secondary Structure Significantly Improves Likelihood - Based Estimates of Phylogeny : Examples from the Bilateria . Molecular Biology and Evolution, 22 0 (4): 0 1129--1136, April 2005. ISSN 07...

    2005 doi

  28. [36]

    Simulations of sequence evolution: how (un) realistic they are and why

    Johanna Trost, Julia Haag, Dimitri H \"o hler, Laurent Jacob, Alexandros Stamatakis, and Bastien Boussau. Simulations of sequence evolution: how (un) realistic they are and why. Molecular Biology and Evolution, 41 0 (1): 0 msad277, 2024

    2024

  29. [37]

    Simplefold: Folding proteins is simpler than you think, 2025

    Yuyang Wang, Jiarui Lu, Navdeep Jaitly, Josh Susskind, and Miguel Angel Bautista. Simplefold: Folding proteins is simpler than you think, 2025. URL https://arxiv.org/abs/2509.18480

    2025

  30. [38]

    Supertree construction: Opportunities and challenges, 2018

    Tandy Warnow. Supertree construction: Opportunities and challenges, 2018. URL https://arxiv.org/abs/1805.03530

    2018 arXiv

  31. [39]

    Minifold: Simple, fast, and accurate protein structure prediction

    Jeremy Wohlwend, Mateo Reveiz, Matt McPartlon, Axel Feldmann, Wengong Jin, and Regina Barzilay. Minifold: Simple, fast, and accurate protein structure prediction. Transactions on Machine Learning Research, 2025. ISSN 2835-8856. URL https://openreview.net/forum?id=1p9hQTbjgo. F...

    2025

  32. [40]

    ART ree: A deep autoregressive model for phylogenetic inference

    Tianyu Xie and Cheng Zhang. ART ree: A deep autoregressive model for phylogenetic inference. In Thirty-seventh Conference on Neural Information Processing Systems, 2023. URL https://openreview.net/forum?id=SoLebIqHgZ

    2023

  33. [41]

    Matsen IV

    Cheng Zhang and Frederick A. Matsen IV. Variational bayesian phylogenetic inference. In International Conference on Learning Representations, 2019. URL https://openreview.net/forum?id=SJVmjjR9FX

    2019

  34. [42]

    Phylo GFN : Phylogenetic inference with generative flow networks

    Ming Yang Zhou, Zichao Yan, Elliot Layne, Nikolay Malkin, Dinghuai Zhang, Moksh Jain, Mathieu Blanchette, and Yoshua Bengio. Phylo GFN : Phylogenetic inference with generative flow networks. In The Twelfth International Conference on Learning Representations, 2024. URL https:/...

    2024

  35. [43]

    Deep residual neural networks resolve quartet molecular phylogenies

    Zhengting Zou, Hongjiu Zhang, Yuanfang Guan, and Jianzhi Zhang. Deep residual neural networks resolve quartet molecular phylogenies. Molecular biology and evolution, 37 0 (5): 0 1495--1507, 2020

    2020

  36. [44]

    Tamuri, Matteo Battini, Fabrícia F

    Sandra Álvarez Carretero, Asif U. Tamuri, Matteo Battini, Fabrícia F. Nascimento, Emily Carlisle, Robert J. Asher, Ziheng Yang, Philip C. J. Donoghue, and Mario dos Reis. A species-level timeline of mammal evolution integrating phylogenomic data. Nature, 602 0 (7896): 0 263--2...

    2022 doi

Pith tools

Reviewed August 4, 2026 · model on record in the stance chip above.