Skip to main content
NCBI home page
Bioinformatics logoLink to Bioinformatics
. 2026 Jun 16;42(7):btag393. doi: 10.1093/bioinformatics/btag393

PhyloNaP: a user-friendly database of phylogeny for natural product–producing enzymes

Aleksandra Korenskaia 1, Martina Adamek 2,3, Judit Szenei 4, Lisa Vader 5, Kai Blin 6, Tillmann Weber 7, Nadine Ziemert 8,9,✉
Editor: Janet Kelso
PMCID: PMC13344837  PMID: 42302386

Abstract

Summary

Phylogenetic analysis is widely used to predict enzyme function, yet building annotated and reusable trees is labor-intensive and requires extensive knowledge about the specific enzymes. Existing resources rarely cover biosynthetic enzymes and lack the context needed for meaningful analysis. We present PhyloNaP, the first large-scale resource dedicated to phylogenies of biosynthetic enzymes. PhyloNaP provides ∼51 000 annotated and interactive trees enriched with chemical, functional, and taxonomic information. Users can classify their own sequences via phylogenetic placement, enabling functional inference in an evolutionary context. A contribution portal allows the community to submit curated trees. By combining scale, breadth of annotation, and interactive functionality, PhyloNaP fills a major gap in bioinformatics resources for enzyme discovery and annotation, with immediate applications to secondary metabolism and beyond.

Availability and implementation

Freely available on the web at https://phylonap.cs.uni-tuebingen.de.

Graphical abstract

Graphical Abstract.

For image description, please refer to the figure legend and surrounding text.

Open in a new tab

1 Introduction

Phylogenetic analysis is a powerful approach for predicting enzyme function, offering insights that go beyond what can be inferred from sequence similarity alone. By placing proteins into an evolutionary context, phylogenies help delineate functional subgroups, identify catalytic mechanisms, and uncover novel biochemistry. This makes phylogeny a widely used strategy in enzyme discovery, biocatalyst engineering, and metabolic pathway reconstruction.

This challenge is particularly acute for enzymes and proteins involved in the biosynthesis of natural products, which are typically encoded in biosynthetic gene clusters (BGCs). Core biosynthetic enzymes such as polyketide synthases (PKSs) and nonribosomal peptide synthetases (NRPSs) can not only be readily detected through conserved domains, but in many cases their modular organization also enables reliable predictions of the core scaffold they assemble (Terlouw et al. 2025). By contrast, tailoring enzymes–including halogenases, glycosyltransferases, oxidoreductases, and methyltransferases–pose a different problem. Their broad enzyme class can usually be assigned from conserved sequence motifs, but it is far more difficult to predict their precise biochemical role: which substrate they act on, the position of a modification, and the type of chemical transformation they perform. These details are crucial, as tailoring reactions define the structural diversity and biological activity of the final compounds (Walsh 2023). In such cases, phylogenetic placement against experimentally characterized homologs provides a more fine-grained basis for functional inference than domain annotation alone (Adamek et al. 2019).

Several widely used genome mining tools have adopted phylogeny-based strategies to improve functional annotation. For example, NaPDoS classifies ketosynthase (KS) and condensation (C) domains based on curated phylogenetic trees to support the identification of PKS and NRPS enzymes (Ziemert et al. 2012). antiSMASH incorporates phylogenetic models and profile Hidden Markov Models (pHMMs) to predict substrate specificities and BGC product classes (Blin et al. 2025). However, these tools serve different purposes: NaPDoS focuses on phylogeny-based classification of selected PKS/NRPS domains, whereas antiSMASH is aimed at detection and annotation of biosynthetic gene clusters. Consequently, a gap remains for a resource that provides reusable, annotated phylogenetic trees for biosynthetic enzymes in a broader functional and chemical context. Additionally, building high-quality phylogenies remains labor-intensive: it requires assembling training sets, aligning sequences, inferring trees, and annotating them with functional or chemical data. Published trees are typically static images that cannot be reused for classifying new sequences. More general phylogenetic databases, such as PhylomeDB (Fuentes et al. 2022) or EggNOG (Hernández-Plaza et al. 2023), provide broad coverage but underrepresent biosynthetic enzymes and lack integration with chemical or genomic context, limiting their utility for this application.

To address these limitations, we developed PhyloNaP, a web-based platform and database of annotated phylogenetic trees for biosynthetic enzymes. PhyloNaP currently hosts ∼51 000 trees, integrating chemical, functional, and taxonomic metadata from multiple sources. The resource supports interactive exploration of trees with customisable metadata display and phylogenetic placement of user-submitted sequences. A contribution portal allows researchers to submit manually annotated trees, ensuring continuous growth supported by the research community.

By bridging the gap between phylogenetic analysis and functional annotation, PhyloNaP provides a community-driven resource of annotated biosynthetic enzyme phylogenies that integrates chemical and functional metadata, connects to complementary databases such as MITE (Rutz et al. 2026), and facilitates applications ranging from structure prediction to enzyme engineering in synthetic biology.

2 Database content and construction

PhyloNaP currently hosts approximately 51 000 phylogenetic trees, each representing a distinct protein family or subfamily involved in natural product biosynthesis (see Fig. 1A). Each dataset comprises a rooted phylogenetic tree in Newick format, a multiple sequence alignment, a FASTA file of untrimmed sequences, and a metadata file with functional, taxonomic, and/or chemical annotations. The majority of sequences are of bacterial origin, with the function ranging from scaffold-forming synthases to diverse tailoring enzymes and other proteins that can be found in BGCs, such as transporters, regulatory and resistance proteins (see Fig. 1B), notably, PhyloNaP covers approximately 80% of tailoring enzymes annotated in MiBiG (see Fig. 1C). Where available, datasets are linked to experimentally characterized reactions and validated BGCs, allowing chemical structures and reaction information to be visualized directly on the tree. For the many proteins that remain uncharacterized, PhyloNaP adds value by providing contextual information—such as BGC type and taxonomy—that guides hypothesis generation and exploratory analysis. Cross-links to external resources further connect each entry to the broader bioinformatics ecosystem and will grow richer as experimental data accumulates.

Figure 1.

Dataset sizes show a broad distribution with a median of 25 sequences. Biosynthetic enzymes represent 28.7 percent of the annotated enzyme datasets in PhyloNaP. Approximately 80 percent of MIBiG functional categories for tailoring enzymes and precursor biosynthesis are represented in the database. The interface displays a phylogenetic tree with multiple annotation panels.

Open in a new tab

(A) Dataset size distribution. (B) Functional overview obtained with eggNOG-mapper (Cantalapiedra et al. 2021, Hernández-Plaza et al. 2023); see details in the Supplementary Materials, available as supplementary data at Bioinformatics online. (C) The overview of the MIBiG functional categories coverage; see details in Supplementary. (D) Tree viewer interface of PhyloNaP.

2.1 Curated and automated datasets

PhyloNaP distinguishes between curated and automatically generated datasets. The curated datasets comprise ten enzyme families that have been manually assembled and annotated by experts, many of them derived from published studies (Reitz et al. 2019, Zdouc et al. 2021, Gavriilidou et al. 2023, Biermann et al. 2024). These serve as high-confidence reference trees for functional classification.

The majority of datasets are generated through an automated pipeline (see Supplementary Materials for details, available as supplementary data at Bioinformatics online) that integrates protein sequences and annotations from established reference databases, clusters them into families with MMseqs2 (Steinegger and Söding 2018), and then consistently aligns with MAFFT (Katoh and Standley 2013), trims with trimAl (Capella-Gutiérrez et al. 2009), and infers rooted trees with FastTree and custom scripts (Bryant and Charleston 2018, Price et al. 2010) suitable for exploration and sequence placement, with additional alignment quality filtering and subclustering of overly diverse datasets (Balaban et al. 2019, Wong et al. 2020). Each dataset is further classified into enzyme superfamilies using HMMER with Superfamily 4.0 profiles (Pandurangan et al. 2019). This workflow enables large-scale coverage while maintaining consistent quality; you can find verification of this approach against published trees in Supplementary, available as supplementary data at Bioinformatics online (Grocholski et al. 2019, Thomy et al. 2019, Hansen et al. 2023, Harper et al. 2024). With ongoing community contributions of curated trees, PhyloNaP is designed to evolve into a comprehensive resource that combines expert knowledge with automated scalability.

3 Classifying user sequences

A key feature of PhyloNaP is its ability to classify user-submitted sequences by placing them onto existing phylogenetic trees. This workflow allows researchers to interpret their sequences in an evolutionary context and infer likely functions or biochemical roles.

Users begin by submitting one or more protein sequences in FASTA format. The system first uses MMseqs2 (Steinegger and Söding 2017) to identify similar proteins within the PhyloNaP database, using a default minimum threshold of 30% identity and 50% coverage; sequences below this threshold are not considered, as they are unlikely to support reliable functional inference. For each matching reference tree, the query sequence is added to the corresponding alignment, and phylogenetic placement is performed (Barbera et al. 2019), usually within seconds or minutes, to determine the most likely positions of the query sequence along the tree. To ensure that placement works correctly we tested PhyloNaP extensively on published examples (Grocholski et al. 2019, Thomy et al. 2019, Hansen et al. 2023, Harper et al. 2024), showed in Supplementary, available as supplementary data at Bioinformatics online.

4 Web interface and user tools

We designed PhyloNaP with a strong emphasis on usability, creating an interface that allows researchers to both explore the database and analyze their own sequences in a single environment. The entry point is a database browser, where users can search and filter thousands of annotated phylogenies by enzyme family, dataset origin (curated or automated), and annotation type. This makes it straightforward to identify relevant reference trees for a particular research question.

To support functional inference, PhyloNaP offers a dedicated Analyze page where users can upload protein sequences in FASTA format. The system automatically identifies the most similar datasets, performs phylogenetic placement, and reports results in a concise table. Each entry provides a description of the matched dataset, quantitative placement scores such as the pendant length (Matsen et al. 2012), and direct links to an interactive tree viewer.

The interactive viewer is central to PhyloNaP. It provides dynamic exploration of phylogenetic trees with features such as rerooting and adjustable branch spacing. Multiple layers of metadata, including functional predictions, taxonomic information, BGC classifications, and, when available, chemical structures or reaction schemes, can be overlaid directly on the tree (Fig. 1C). These annotations are fully interactive. Structures can be expanded for inspection, and hyperlinks lead directly to external resources such as MiBiG (Zdouc et al. 2024), which links experimentally validated BGCs to their products; SwissProt (Boutet et al. 2007, The UniProt Consortium 2025, Bateman et al. 2025), providing curated protein sequences with functional data; MITE (Zdouc et al. 2024), which contributes manually curated reaction data for tailoring enzymes; and PanBGC (Paccagnella et al. 2025) enabling comparative analysis of similar BGCs from the antiSMASH database (Blin et al. 2024). Together, these links provide users with direct access to high-quality chemical, functional, and genomic context. When user-submitted sequences are placed, they appear as highlighted nodes with evolutionary distance visualized by color gradient. An annotation summary of the most recent common ancestor of all placement nodes is displayed by default, providing an immediate overview of the functional and genomic context.

Beyond browsing and analysis of phylogenetic trees, PhyloNaP also supports community engagement. A submission portal enables researchers to contribute their own curated datasets, which are reviewed for quality before being integrated into the public resource. Each submission must include a rooted phylogenetic tree, the corresponding multiple sequence alignment, untrimmed protein sequences, and a metadata table. Detailed guidelines and recommended additional information are provided on the Help page. This feature ensures that the database continues to grow not only in scale but also in depth, combining expert knowledge with automated coverage.

As an illustration, we tested the placement of a flavin-dependent halogenase (A0A1L1QK36 in UniProt) (Fig. 1C). The analysis page reports placements onto two reference datasets, with the prioritized tree showing a high LWR (0.998) and a short pendant length (0.04), or the evolutionary distance from the node of placement. The sequence is positioned in a clade dominated by proteins from NRPS BGCs, suggesting amino acid or peptide substrates. One closely related enzyme is annotated as a tryptophan-6 halogenase, and the natural product structures linked to this and neighboring clades consistently carry a halogen substituent at this position. Taken together, the reliable positioning, very short branch length, and phylogenetic and chemical evidence strongly suggest that the query protein performs a tryptophan-6 halogenation This example demonstrates how PhyloNaP’s interface integrates evolutionary, functional, and chemical information to support detailed functional predictions; a more detailed description of the workflow is provided in the Supplementary Materials, available as supplementary data at Bioinformatics online.

5 Conclusion and discussion

PhyloNaP establishes a centralized platform for annotated phylogenetic trees of biosynthetic enzymes. By integrating functional, chemical, and taxonomic metadata into interactive phylogenies, it enables researchers to investigate enzyme families in an evolutionary framework and to classify novel sequences with confidence.

A key strength of PhyloNaP lies in its integration with external resources such as MiBiG, MITE, SwissProt, and PanBGC. These links enrich the evolutionary trees with experimentally validated functions, reaction data, and genomic context, allowing users to connect sequence placement directly to biochemical and structural information. Such integration facilitates practical applications ranging from structure prediction of natural products to the identification of candidate enzymes for pathway engineering in synthetic biology.

The resource is not without limitations. Automated tree rooting may be suboptimal for some families, although users can interactively reroot trees for visualization; incorrect rooting may affect interpretation of deep (basal) placements, where manual verification of root position is recommended prior to functional inference. Functional annotations depend on the current state of reference databases, and many biosynthetic enzymes remain poorly characterized, meaning that predictions often reflect context rather than exact reaction specificity. In addition, sequence placement is currently limited to 50 sequences to ensure computational efficiency, and manual inspection remains essential for reliable interpretation.

Looking ahead, PhyloNaP is designed as a living, community-driven resource. Regular updates to MiBiG and MITE will steadily expand the pool of experimentally characterized proteins, while direct user contributions will add expert-curated phylogenies. Future development will focus on tighter integration with genome-mining and machine-learning tools, deeper curation of submitted datasets, and expanded coverage of published trees. By reducing redundancy, improving accessibility, and fostering community input, PhyloNaP has the potential to accelerate functional annotation, enable discovery of new enzymatic activities, and support the rational exploration of biosynthetic diversity.

Supplementary Material

btag393_Supplementary_Data
btag393_supplementary_data.pdf (3.3MB, pdf)

Acknowledgements

We would like to acknowledge Noel Kubach, Dr. Mitja Zdouc, Mathiaz Witte Paz, and Konstantin Vysotskii for discussions and advice on the tool development; and Dr. Leo Padva and Dr. Athina Gavriilidou for sharing datasets.

Contributor Information

Aleksandra Korenskaia, Translational Genome Mining for Natural Products, Interfaculty Institute of Microbiology and Infection Medicine Tübingen (IMIT), Interfaculty Institute for Biomedical Informatics (IBMI), University of Tübingen, Tübingen, 72076, Germany.

Martina Adamek, Translational Genome Mining for Natural Products, Interfaculty Institute of Microbiology and Infection Medicine Tübingen (IMIT), Interfaculty Institute for Biomedical Informatics (IBMI), University of Tübingen, Tübingen, 72076, Germany; German Center for Infection Research (DZIF), Tübingen, 72076, Germany.

Judit Szenei, Department of Biotechnology and Biomedicine, Technical University of Denmark, Kgs. Lyngby, Denmark.

Lisa Vader, Department of Biotechnology and Biomedicine, Technical University of Denmark, Kgs. Lyngby, Denmark.

Kai Blin, Department of Biotechnology and Biomedicine, Technical University of Denmark, Kgs. Lyngby, Denmark.

Tillmann Weber, Department of Biotechnology and Biomedicine, Technical University of Denmark, Kgs. Lyngby, Denmark.

Nadine Ziemert, Translational Genome Mining for Natural Products, Interfaculty Institute of Microbiology and Infection Medicine Tübingen (IMIT), Interfaculty Institute for Biomedical Informatics (IBMI), University of Tübingen, Tübingen, 72076, Germany; German Center for Infection Research (DZIF), Tübingen, 72076, Germany.

Author contributions

Aleksandra Korenskaia (Conceptualization [Equal], Data curation [Lead], Formal analysis [Lead], Investigation [Lead], Methodology [Lead], Software [Lead], Validation [Lead], Visualization [Lead], Writing—original draft [Lead], Writing—review & editing [Equal]), Martina Adamek (Validation [Lead], Writing—review & editing [Equal]), Judit Szenei (Data curation [Supporting], Validation [Supporting], Writing—review & editing [Supporting]), Lisa Vader (Data curation [Supporting], Formal analysis [Supporting], Writing—review & editing [Supporting]), Kai Blin (Supervision [Supporting], Writing—review & editing [Supporting]), Tilmann Weber (Funding acquisition [Supporting], Writing—review & editing [Supporting]), and Nadine Ziemert (Conceptualization [Equal], Funding acquisition [Lead], Project administration [Lead], Resources [Lead], Supervision [Lead], Writing—original draft [Lead], Writing—review & editing [Lead])

Supplementary material

Supplementary material is available at Bioinformatics online.

Conflict of interest

None declared.

Funding

This work was supported by the European Union’s Horizon Europe program [Marie Skłodowska-Curie grant agreement No. 101072485 (MAGic-MOLFUN)]; the Novo Nordisk Foundation [NNF20CC0035580 to T.W.]; the Federal Ministry of Education and Research (BMBF)[N.Z]; the German Centre for Infection Research (DZIF) [TTU09.716 to M.A. and N.Z.]; and the Cluster of Excellence “Controlling Microbes to Fight Infection” (CMFI, project ID 390838134, EXC 2124) [structural support to A.K. and N.Z.].

Data availability

The PhyloNaP database is available for browsing and downloading individual datasets at https://phylonap.cs.uni-tuebingen.de, and the full database can be downloaded from https://phylonap.cs.uni-tuebingen.de/download. The results of the functional distribution analysis reported in this article are provided as Supplementary Table, available as supplementary data at Bioinformatics online. The source code of the PhyloNaP web application is available at https://github.com/ZiemertLab/PhyloNaP_WebApp and DOI: 10.5281/zenodo.19541951. The database generation workflow is available at https://github.com/ZiemertLab/PhyloNaP_createdb, DOI: 10.5281/zenodo.1956716.

References

  1. Adamek M, Alanjary M, Ziemert N.  Applied evolution: phylogeny-based approaches in natural products research. Nat Prod Rep  2019;36:1295–312. 10.1039/C9NP00027E. [DOI] [PubMed] [Google Scholar]
  2. Balaban M, Moshiri N, Mai U  et al.  TreeCluster: clustering biological sequences using phylogenetic trees. PLoS One  2019;14:e0221068. 10.1371/journal.pone.0221068. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Barbera P, Kozlov AM, Czech L  et al.  EPA-Ng: massively parallel evolutionary placement of genetic sequences. Syst Biol  2019;68:365–9. 10.1093/sysbio/syy054. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bateman A, Martin M-J, Orchard S  et al.  UniProt: the universal protein knowledgebase in 2025. Nucleic Acids Res  2025;53:D609–17. 10.1093/nar/gkae1010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Biermann F, Tan B, Breitenbach M  et al.  Exploration, expansion and definition of the atropopeptide family of ribosomally synthesized and posttranslationally modified peptides. Chem Sci  2024;15:17506–23. 10.1039/D4SC03469D. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Blin K, , ShawS, , Medema MHet al. The antiSMASH database version 4: additional genomes and BGCs, new sequence-based searches and more. Nucleic Acids Res  2024;52:D586–9. 10.1093/nar/gkad984. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Blin K, , ShawS, , Vader Let al. Antismash 8.0: extended gene cluster detection capabilities and analyses of chemistry, enzymology, and regulation. Nucleic Acids Res  2025;53:W32–8. 10.1093/nar/gkaf334. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Boutet E, Lieberherr D, Tognolli M  et al.  UniProtKB/Swiss-Prot. Methods Mol Biol (Clifton, N.J.)  2007;406:89–112. 10.1007/978-1-59745-535-0_4. [DOI] [PubMed] [Google Scholar]
  9. Bryant D, Charleston M. MAD roots for large trees. arXiv : 1811.03174. Preprint, arXiv. 10.48550/arXiv.1811.03174, 7 November 2018, preprint: not peer reviewed. [DOI]
  10. Cantalapiedra CP, Hernández-Plaza A, Letunic I  et al.  eggNOG-Mapper v2: functional annotation, orthology assignments, and domain prediction at the metagenomic scale. Mol Biol Evol  2021;38:5825–9. 10.1093/molbev/msab293. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Capella-Gutiérrez S, Silla-Martínez JM, Gabaldón T.  trimAl: a tool for automated alignment trimming in Large-Scale phylogenetic analyses. Bioinformatics  2009;25:1972–3. 10.1093/bioinformatics/btp348. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Fuentes D, Molina M, Chorostecki U  et al.  PhylomeDB V5: an expanding repository for genome-wide catalogues of annotated gene phylogenies. Nucleic Acids Res  2022;50:D1062–8. 10.1093/nar/gkab966. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Gavriilidou A, Adamek M, Rodler J-P  et al. Phylogenetic distance and structural diversity directing a reclassification of glycopeptide antibiotics. bioRxiv, 10.1101/2023.02.10.526856, 11 February 2023, preprint. [DOI]
  14. Grocholski T, , YamadaK, , Sinkkonen Jet al. Evolutionary trajectories for the functional diversification of anthracycline methyltransferases. ACS Chem Biol  2019;14:850–6. 10.1021/acschembio.9b00238. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Hansen MH, Adamek M, Iftime D  et al.  Resurrecting ancestral antibiotics: unveiling the origins of modern lipid II targeting glycopeptides. Nat Commun  2023;14:7842. 10.1038/s41467-023-43451-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Harper CP, Day A, Tsingos M  et al.  Critical analysis of polycyclic tetramate macrolactam biosynthetic gene cluster phylogeny and functional diversity. Appl Environ Microbiol  2024;90:e00600–24. 10.1128/aem.00600-24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Hernández-Plaza A, Szklarczyk D, Botas J  et al.  eggNOG 6.0: enabling comparative genomics across 12 535 organisms. Nucleic Acids Res  2023;51:D389–94. 10.1093/nar/gkac1022. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Katoh K, Standley DM.  MAFFT multiple sequence alignment software version 7: improvements in performance and usability. Mol Biol Evol  2013;30:772–80. 10.1093/molbev/mst010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Matsen FA, Hoffman NG, Gallagher A  et al.  A format for phylogenetic placements. PLoS One  2012;7:e31009. 10.1371/journal.pone.0031009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Paccagnella D, Bagci C, Gavriilidou A  et al. PanBGC: a pangenome-inspired framework for comparative analysis of biosynthetic gene clusters. bioRxiv, 10.1101/2025.08.11.669102, 11 August 2025, preprint. [DOI] [PMC free article] [PubMed]
  21. Pandurangan AP, Stahlhacke J, Oates ME  et al.  The SUPERFAMILY 2.0 database: a significant proteome update and a new webserver. Nucleic Acids Research  2019;47:D490–94. 10.1093/nar/gky1130. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Price MN, Dehal PS, Arkin AP.  FastTree 2—approximately maximum-likelihood trees for large alignments. PLoS One  2010;5:e9490. 10.1371/journal.pone.0009490. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Reitz ZL, Hardy CD, Suk J  et al.  Genomic analysis of siderophore β-hydroxylases reveals divergent stereocontrol and expands the condensation domain family. Proc Natl Acad Sci USA  2019;116:19805–14. 10.1073/pnas.1903161116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Rutz A, , ProbstD, , Aguilar Cet al. Mite: the minimum information about a tailoring enzyme database for capturing specialized metabolite biosynthesis. Nucleic Acids Res  2026;54:D635–42. 10.1093/nar/gkaf969. 41001881. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Steinegger M, Söding J.  Clustering huge protein sequence sets in linear time. Nat Commun  2018;9:2542. 10.1038/s41467-018-04964-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Steinegger M, Söding J.  MMseqs2 enables sensitive protein sequence searching for the analysis of massive data sets. Nat Biotechnol  2017;35:1026–8. 10.1038/nbt.3988. [DOI] [PubMed] [Google Scholar]
  27. Terlouw BR, Huang C, Meijer D  et al. PARAS: high-accuracy machine-learning of substrate specificities in nonribosomal peptide synthetases. bioRxiv, 10.1101/2025.01.08.631717, 10 January 2025, preprint. [DOI] [PMC free article] [PubMed]
  28. The UniProt Consortium. 2025. “UniProt: The Universal Protein Knowledgebase in 2025.” Nucleic Acids Res 53: D609–17. 10.1093/nar/gkae1010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Thomy D, Culp E, Adamek M  et al.  The ADEP biosynthetic gene cluster in Streptomyces hawaiiensis NRRL 15010 reveals an accessory clpP gene as a novel antibiotic resistance factor. Appl Environ Microbiol  2019;85:e01292-19. 10.1128/AEM.01292-19. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Walsh CT.  Tailoring enzyme strategies and functional groups in biosynthetic pathways. Nat Prod Rep  2023;40:326–86. 10.1039/D2NP00048B. [DOI] [PubMed] [Google Scholar]
  31. Wong TKF, Kalyaanamoorthy S, Meusemann K  et al.  A minimum reporting standard for multiple sequence alignments. NAR Genom Bioinform  2020;2:lqaa024. 10.1093/nargab/lqaa024. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Zdouc MM, Alanjary MM, Zarazúa GS  et al.  A biaryl-linked tripeptide from planomonospora reveals a widespread class of minimal RiPP gene clusters. Cell Chem Biol  2021;28:733–9.e4. 10.1016/j.chembiol.2020.11.009. [DOI] [PubMed] [Google Scholar]
  33. Zdouc MM, Meijer D, Biermann F  et al. The minimum information about a tailoring enzyme/maturase data standard for capturing natural product biosynthesis. 10.26434/chemrxiv-2024-78mtl, ChemRxiv, 9 April 2024, preprint. [DOI]
  34. Zdouc MM, Blin K, Louwen NLL  et al.  MIBiG 4.0: advancing biosynthetic gene cluster curation through global collaboration. Nucleic Acids Res  2025;53:gkae1115–D690. 10.1093/nar/gkae1115. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Ziemert N, Podell S, Penn K  et al.  The natural product domain seeker NaPDoS: a phylogeny based bioinformatic tool to classify secondary metabolite gene diversity. PLoS One  2012;7:e34064. 10.1371/journal.pone.003406. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

btag393_Supplementary_Data
btag393_supplementary_data.pdf (3.3MB, pdf)

Data Availability Statement

The PhyloNaP database is available for browsing and downloading individual datasets at https://phylonap.cs.uni-tuebingen.de, and the full database can be downloaded from https://phylonap.cs.uni-tuebingen.de/download. The results of the functional distribution analysis reported in this article are provided as Supplementary Table, available as supplementary data at Bioinformatics online. The source code of the PhyloNaP web application is available at https://github.com/ZiemertLab/PhyloNaP_WebApp and DOI: 10.5281/zenodo.19541951. The database generation workflow is available at https://github.com/ZiemertLab/PhyloNaP_createdb, DOI: 10.5281/zenodo.1956716.

Articles from Bioinformatics are provided here courtesy of Oxford University Press

RESOURCES