{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,28]],"date-time":"2026-01-28T02:10:11Z","timestamp":1769566211044,"version":"3.49.0"},"reference-count":17,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2025,12,23]],"date-time":"2025-12-23T00:00:00Z","timestamp":1766448000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,1,27]],"date-time":"2026-01-27T00:00:00Z","timestamp":1769472000000},"content-version":"vor","delay-in-days":35,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"Max Planck Institute for Multidisciplinary Sciences"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["BMC Bioinformatics"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Most popular tools for reconstructing phylogenetic trees from multiple sequence alignments use a model of molecular evolution in which a single substitution matrix or a small set of fixed matrices are shared between all columns. Models with column-specific rate matrices can in principle be fit by automatic differentiation methods, but in practice the heavy computational burden associated with computing the gradients of the many matrix exponentials has hindered exploration of such models.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Implementation<\/jats:title>\n                    <jats:p>Here, we present a highly efficient approach for reverse-mode differentiation of the log likelihood computed with Felsenstein\u2019s algorithm under any time-reversible substitution model. PhyloGrad is implemented in Rust and has Python bindings to easily combine it with automatic differentiation tools.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Depending on the tree size, PhyloGrad is 30-100 times faster than automatic differentiation in Pytorch and uses 10-100 times less memory. Even in the task of fitting one global model it is still at least 10 times faster than IQ-TREE3. PhyloGrad accelerates current model optimizations and enables the field to easily explore and implement novel site-specific models.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1186\/s12859-025-06353-4","type":"journal-article","created":{"date-parts":[[2025,12,23]],"date-time":"2025-12-23T06:47:59Z","timestamp":1766472479000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Phylograd: fast column-specific calculation of substitution model gradients"],"prefix":"10.1186","volume":"27","author":[{"given":"Benjamin","family":"Lieser","sequence":"first","affiliation":[]},{"given":"Georgy","family":"Belousov","sequence":"additional","affiliation":[]},{"given":"Johannes","family":"S\u00f6ding","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2025,12,23]]},"reference":[{"key":"6353_CR1","doi-asserted-by":"publisher","first-page":"1229","DOI":"10.1111\/j.1558-5646.1981.tb04991.x","volume":"35","author":"J Felsenstein","year":"1981","unstructured":"Felsenstein J. Evolutionary trees from gene frequencies and quantitative characters: finding maximum likelihood estimates. Evolution. 1981;35:1229\u201342.","journal-title":"Evolution"},{"key":"6353_CR2","doi-asserted-by":"crossref","unstructured":"Wong TK, Ly-Trong N, Ren H, Ba\u00f1os H, Roger AJ, Susko E, Bielow C, De\u00a0Maio N, Goldman N, Hahn MW, et al. IQ-TREE 3: phylogenomic inference software using complex evolutionary models. 2025.","DOI":"10.32942\/X2P62N"},{"issue":"21","key":"6353_CR3","doi-asserted-by":"publisher","first-page":"4453","DOI":"10.1093\/bioinformatics\/btz305","volume":"35","author":"AM Kozlov","year":"2019","unstructured":"Kozlov AM, Darriba D, Flouri T, Morel B, Stamatakis A. RAxML-NG: a fast, scalable and user-friendly tool for maximum likelihood phylogenetic inference. Bioinformatics. 2019;35(21):4453\u20135.","journal-title":"Bioinformatics"},{"issue":"7","key":"6353_CR4","doi-asserted-by":"publisher","first-page":"1307","DOI":"10.1093\/molbev\/msn067","volume":"25","author":"SQ Le","year":"2008","unstructured":"Le SQ, Gascuel O. An improved general amino acid replacement matrix. Mol Biol Evol. 2008;25(7):1307\u201320. https:\/\/doi.org\/10.1093\/molbev\/msn067.","journal-title":"Mol Biol Evol"},{"issue":"5","key":"6353_CR5","doi-asserted-by":"publisher","first-page":"691","DOI":"10.1093\/oxfordjournals.molbev.a003851","volume":"18","author":"S Whelan","year":"2001","unstructured":"Whelan S, Goldman N. A general empirical model of protein evolution derived from multiple protein families using a maximum-likelihood approach. Mol Biol Evol. 2001;18(5):691\u20139. https:\/\/doi.org\/10.1093\/oxfordjournals.molbev.a003851.","journal-title":"Mol Biol Evol"},{"issue":"20","key":"6353_CR6","doi-asserted-by":"publisher","first-page":"2317","DOI":"10.1093\/bioinformatics\/btn445","volume":"24","author":"L Si Quang","year":"2008","unstructured":"Si Quang L, Gascuel O, Lartillot N. Empirical profile mixture models for phylogenetic reconstruction. Bioinformatics. 2008;24(20):2317\u201323. https:\/\/doi.org\/10.1093\/bioinformatics\/btn445.","journal-title":"Bioinformatics"},{"issue":"6","key":"6353_CR7","doi-asserted-by":"publisher","first-page":"1095","DOI":"10.1093\/molbev\/msh112","volume":"21","author":"N Lartillot","year":"2004","unstructured":"Lartillot N, Philippe H. A bayesian mixture model for across-site heterogeneities in the amino-acid replacement process. Mol Biol Evol. 2004;21(6):1095\u2013109. https:\/\/doi.org\/10.1093\/molbev\/msh112.","journal-title":"Mol Biol Evol"},{"issue":"9","key":"6353_CR8","doi-asserted-by":"publisher","first-page":"174","DOI":"10.1093\/molbev\/msae174","volume":"41","author":"H Ba\u00f1os","year":"2024","unstructured":"Ba\u00f1os H, Wong TK, Daneau J, Susko E, Minh BQ, Lanfear R, et al. Gtrpmix: a linked general time-reversible model for profile mixture models. Mol Biol Evol. 2024;41(9):174.","journal-title":"Mol Biol Evol"},{"issue":"1","key":"6353_CR9","doi-asserted-by":"publisher","first-page":"503","DOI":"10.1007\/BF01589116","volume":"45","author":"DC Liu","year":"1989","unstructured":"Liu DC, Nocedal J. On the limited memory BFGS method for large scale optimization. Math Program. 1989;45(1):503\u201328.","journal-title":"Math Program"},{"issue":"10","key":"6353_CR10","doi-asserted-by":"publisher","first-page":"3047","DOI":"10.1093\/molbev\/msaa130","volume":"37","author":"X Ji","year":"2020","unstructured":"Ji X, Zhang Z, Holbrook A, Nishimura A, Baele G, Rambaut A, et al. Gradients do grow on trees: a linear-time o(n)-dimensional gradient for statistical phylogenetics. Mol Biol Evol. 2020;37(10):3047\u201360. https:\/\/doi.org\/10.1093\/molbev\/msaa130.","journal-title":"Mol Biol Evol"},{"issue":"3","key":"6353_CR11","doi-asserted-by":"publisher","DOI":"10.1063\/1.4928596","volume":"143","author":"RT McGibbon","year":"2015","unstructured":"McGibbon RT, Pande VS. Efficient maximum likelihood parameterization of continuous-time markov processes. J Chem Phys. 2015;143(3):034109. https:\/\/doi.org\/10.1063\/1.4928596.","journal-title":"J Chem Phys."},{"issue":"3","key":"6353_CR12","doi-asserted-by":"publisher","first-page":"970","DOI":"10.1137\/09074721X","volume":"31","author":"AH Al-Mohy","year":"2010","unstructured":"Al-Mohy AH, Higham NJ. A new scaling and squaring algorithm for the matrix exponential. SIAM J Matrix Anal Appl. 2010;31(3):970\u201389. https:\/\/doi.org\/10.1137\/09074721X.","journal-title":"SIAM J Matrix Anal Appl"},{"issue":"3","key":"6353_CR13","doi-asserted-by":"publisher","first-page":"679","DOI":"10.1137\/0725041","volume":"25","author":"CD Meyer","year":"1988","unstructured":"Meyer CD, Stewart GW. Derivatives and perturbations of eigenvectors. SIAM J Numer Anal. 1988;25(3):679\u201391.","journal-title":"SIAM J Numer Anal"},{"issue":"3","key":"6353_CR14","doi-asserted-by":"publisher","first-page":"321","DOI":"10.1006\/aama.1995.1015","volume":"16","author":"I Najfeld","year":"1995","unstructured":"Najfeld I, Havel TF. Derivatives of the matrix exponential and their computation. Adv Appl Math. 1995;16(3):321\u201375. https:\/\/doi.org\/10.1006\/aama.1995.1015.","journal-title":"Adv Appl Math"},{"issue":"5","key":"6353_CR15","doi-asserted-by":"publisher","first-page":"1046","DOI":"10.1093\/sysbio\/syab010","volume":"70","author":"BQ Minh","year":"2021","unstructured":"Minh BQ, Dang CC, Vinh LS, Lanfear R. Qmaker: fast and accurate method to estimate empirical models of protein evolution. Syst Biol. 2021;70(5):1046\u201360. https:\/\/doi.org\/10.1093\/sysbio\/syab010.","journal-title":"Syst Biol"},{"key":"6353_CR16","doi-asserted-by":"publisher","DOI":"10.1002\/9781118723203","volume-title":"Practical methods of optimization","author":"R Fletcher","year":"2000","unstructured":"Fletcher R. Practical methods of optimization. Hoboken: John Wiley & Sons; 2000."},{"key":"6353_CR17","doi-asserted-by":"crossref","unstructured":"Ji X, Redelings B, Su S, Bao H, Deng W-M, Hong SL, Baele G, Lemey P, Suchard MA. Detecting evolutionary change-points with branch-specific substitution models and shrinkage priors. arXiv preprint arXiv:2507.08386 2025.","DOI":"10.21203\/rs.3.rs-6926809\/v1"}],"container-title":["BMC Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s12859-025-06353-4","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12859-025-06353-4.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12859-025-06353-4.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,1,27]],"date-time":"2026-01-27T11:45:57Z","timestamp":1769514357000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1186\/s12859-025-06353-4"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,12,23]]},"references-count":17,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["6353"],"URL":"https:\/\/doi.org\/10.1186\/s12859-025-06353-4","relation":{},"ISSN":["1471-2105"],"issn-type":[{"value":"1471-2105","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,12,23]]},"assertion":[{"value":"4 September 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"11 December 2025","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"23 December 2025","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"Not applicable","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics approval and consent to participate"}},{"value":"Not applicable","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent for publication"}},{"value":"The authors declare no Conflict of interest.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"20"}}