E-Mail: tom.eulenfeld@uni-jena.de
Room: 4006
Phone: +49-3641-9-46484
I am a scientific programmer with a background in physics and seismology. As a side project, I develop sugar, a lightweight Python toolkit for working with DNA/RNA sequences and bioinformatics file formats. Check it out on GitHub along with my other software projects. If you’re curious about my pre-bioinformatics days, older papers are listed on Google Scholar.
Publications
2026
Triebel, Sandra; Lamkiewicz, Kevin; Eulenfeld, Tom; Marz, Manja
Reducing haystacks to needles – ViralClust: A Nextflow pipeline to cluster viral sequences Journal Article
In: bioRxiv, no. 2026.01.30.702815, 2026.
@article{nokey_100,
title = {Reducing haystacks to needles – ViralClust: A Nextflow pipeline to cluster viral sequences},
author = {Sandra Triebel and Kevin Lamkiewicz and Tom Eulenfeld and Manja Marz},
doi = {10.64898/2026.01.30.702815},
year = {2026},
date = {2026-06-11},
journal = {bioRxiv},
number = {2026.01.30.702815},
abstract = {Background The rapid accumulation of viral genome sequences presents major challenges for downstream analysis tools, including tools for multiple sequence alignments, phylogeny, and genome/alignment visualization, due to computational constraints and sampling biases caused by outbreak-driven over-representation. Selecting representative genomes through clustering offers a principled alternative to random subsampling, yet choosing appropriate clustering strategies remains non-trivial and context dependent.
Results Here, we present ViralClust, a modular Nextflow pipeline for bias-aware representative selection from large viral genome datasets. ViralClust integrates five distinct clustering algorithms (CD-HIT-EST, SUMACLUST, VSEARCH, MMSeqs2, and HDBSCAN) within a unified workflow, enabling direct comparison of clustering outcomes and flexible adaptation to diverse biological questions, considering a balanced phylogenic distribution of the selected sequences. We evaluated ViralClust on six RNA and DNA virus datasets ranging from 632 to 156,586 sequences and spanning genome lengths from 890 to 197,185 nucleotides. Across all datasets, clustering reduced dataset size by ∼95 % or more while preserving genetic diversity across species, genera, and families, and effectively mitigating biases introduced by outbreaks, partial genomes, and sequence orientation artifacts.
Conclusions By supporting whole-genome clustering and scalable representative selection, ViralClust enables efficient and reproducible downstream analyses that would otherwise be computationally infeasible. Our framework provides a flexible foundation for large-scale viral genomics and supports future applications in comparative analysis and virus classification.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Background The rapid accumulation of viral genome sequences presents major challenges for downstream analysis tools, including tools for multiple sequence alignments, phylogeny, and genome/alignment visualization, due to computational constraints and sampling biases caused by outbreak-driven over-representation. Selecting representative genomes through clustering offers a principled alternative to random subsampling, yet choosing appropriate clustering strategies remains non-trivial and context dependent.
Results Here, we present ViralClust, a modular Nextflow pipeline for bias-aware representative selection from large viral genome datasets. ViralClust integrates five distinct clustering algorithms (CD-HIT-EST, SUMACLUST, VSEARCH, MMSeqs2, and HDBSCAN) within a unified workflow, enabling direct comparison of clustering outcomes and flexible adaptation to diverse biological questions, considering a balanced phylogenic distribution of the selected sequences. We evaluated ViralClust on six RNA and DNA virus datasets ranging from 632 to 156,586 sequences and spanning genome lengths from 890 to 197,185 nucleotides. Across all datasets, clustering reduced dataset size by ∼95 % or more while preserving genetic diversity across species, genera, and families, and effectively mitigating biases introduced by outbreaks, partial genomes, and sequence orientation artifacts.
Conclusions By supporting whole-genome clustering and scalable representative selection, ViralClust enables efficient and reproducible downstream analyses that would otherwise be computationally infeasible. Our framework provides a flexible foundation for large-scale viral genomics and supports future applications in comparative analysis and virus classification.
Results Here, we present ViralClust, a modular Nextflow pipeline for bias-aware representative selection from large viral genome datasets. ViralClust integrates five distinct clustering algorithms (CD-HIT-EST, SUMACLUST, VSEARCH, MMSeqs2, and HDBSCAN) within a unified workflow, enabling direct comparison of clustering outcomes and flexible adaptation to diverse biological questions, considering a balanced phylogenic distribution of the selected sequences. We evaluated ViralClust on six RNA and DNA virus datasets ranging from 632 to 156,586 sequences and spanning genome lengths from 890 to 197,185 nucleotides. Across all datasets, clustering reduced dataset size by ∼95 % or more while preserving genetic diversity across species, genera, and families, and effectively mitigating biases introduced by outbreaks, partial genomes, and sequence orientation artifacts.
Conclusions By supporting whole-genome clustering and scalable representative selection, ViralClust enables efficient and reproducible downstream analyses that would otherwise be computationally infeasible. Our framework provides a flexible foundation for large-scale viral genomics and supports future applications in comparative analysis and virus classification.
Triebel, Sandra; Eulenfeld, Tom; Ontiveros-Palacios, Nancy; Sweeney, Blake; Tautz, Norbert; Marz, Manja
First full-genome alignment representative for the genus Pestivirus Journal Article
In: RNA, vol. 32, iss. 3, pp. 311-327, 2026.
@article{nokey_77,
title = {First full-genome alignment representative for the genus \textit{Pestivirus}},
author = {Sandra Triebel and Tom Eulenfeld and Nancy Ontiveros-Palacios and Blake Sweeney and Norbert Tautz and Manja Marz},
doi = {10.1261/rna.080732.125},
year = {2026},
date = {2026-03-01},
urldate = {2025-05-27},
journal = {RNA},
volume = {32},
issue = {3},
pages = {311-327},
abstract = {Members of the genus Pestivirus (family Flaviviridae) comprise economically important livestock pathogens like classical swine fever virus (CSFV) and bovine viral diarrhea virus (BVDV). Research over recent years revealed 11 recognized and eight proposed species. The single-stranded, positive-sense RNA genome encodes one large polyprotein that is processed by viral and cellular proteases into 12 mature proteins. In addition to its protein-coding function, the RNA genome contains secondary structures critical for various stages of the viral life cycle. Some of these structures, including the internal ribosome entry site (IRES) and a 3′ stem–loop, essential for genome replication, have been studied in individual pestiviruses. Here, we present the first genome-wide multiple sequence alignment comprising all known pestivirus species (accepted and tentative) and a comprehensive analysis of phylogenetically conserved RNA secondary structures across the genus. Well-characterized elements, such as a 5′ stem–loop, the IRES, and the 3′ stem–loop SL I, were conserved in all pestiviruses, whereas additional 3′ untranslated region structures were conserved only in subsets of species. We identified 29 novel conserved RNA secondary structures within the protein-coding region, with thus far unresolved functional importance. A miR-17 binding site, previously described in species A, B, and C, was detected in ten additional species but absent in species K, S, Q, and R. We identified a putative long-distance RNA interaction between the IRES and the 3′ end of the genome. Together, these findings and the comprehensive MSA of all 19 pestivirus species provide a valuable resource for future research and diagnostic applications.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Members of the genus Pestivirus (family Flaviviridae) comprise economically important livestock pathogens like classical swine fever virus (CSFV) and bovine viral diarrhea virus (BVDV). Research over recent years revealed 11 recognized and eight proposed species. The single-stranded, positive-sense RNA genome encodes one large polyprotein that is processed by viral and cellular proteases into 12 mature proteins. In addition to its protein-coding function, the RNA genome contains secondary structures critical for various stages of the viral life cycle. Some of these structures, including the internal ribosome entry site (IRES) and a 3′ stem–loop, essential for genome replication, have been studied in individual pestiviruses. Here, we present the first genome-wide multiple sequence alignment comprising all known pestivirus species (accepted and tentative) and a comprehensive analysis of phylogenetically conserved RNA secondary structures across the genus. Well-characterized elements, such as a 5′ stem–loop, the IRES, and the 3′ stem–loop SL I, were conserved in all pestiviruses, whereas additional 3′ untranslated region structures were conserved only in subsets of species. We identified 29 novel conserved RNA secondary structures within the protein-coding region, with thus far unresolved functional importance. A miR-17 binding site, previously described in species A, B, and C, was detected in ten additional species but absent in species K, S, Q, and R. We identified a putative long-distance RNA interaction between the IRES and the 3′ end of the genome. Together, these findings and the comprehensive MSA of all 19 pestivirus species provide a valuable resource for future research and diagnostic applications.
2025
Eulenfeld, Tom; Triebel, Sandra; Marz, Manja
AnchoRNA: Full virus genome alignments through conserved anchor regions Journal Article
In: bioRxiv, 2025.
@article{nokey_67,
title = {AnchoRNA: Full virus genome alignments through conserved anchor regions},
author = {Tom Eulenfeld and Sandra Triebel and Manja Marz},
doi = {10.1101/2025.01.30.635689},
year = {2025},
date = {2025-12-15},
urldate = {2025-02-01},
journal = {bioRxiv},
abstract = {Multiple sequence alignment of full viral genomes can be challenging due to factors such as long sequences, large insertions/deletions (spanning several 100 nucleotides), large number of sequences, sequence divergence, and high computational complexity in particular when computing alignments based on RNA secondary structures. Standard alignment methods often face these issues, in particular when processing highly variable sequences or when specific phylogenetic analysis is required on selected subsequences.
We present an algorithm to determine high quality anchors that define partitions of sequences and guide the alignment of viral genomes to respect well conserved, and therefore functionally significant, regions. This new approach is implemented in the Python-based command line tool AnchoRNA, which is designed to identify conserved regions, or anchors, within coding sequences. By default, anchors are searched in translated coding sequences accounting for high mutation rates in viral genomes. AnchoRNA enhances the accuracy and efficiency of full-genome alignment by focusing on these crucial conserved regions. AnchoRNA guided alignments are systematically compared to the results of 3 alignment programs. Utilizing a dataset of 55 representative Pestivirus genomes, AnchoRNA identified 55 anchors that are used for guiding the alignment process. The incorporation of these anchors led to improvements across tested alignment tools, highlighting the effectiveness of AnchoRNA in enhancing alignment quality, especially in viral genomes.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Multiple sequence alignment of full viral genomes can be challenging due to factors such as long sequences, large insertions/deletions (spanning several 100 nucleotides), large number of sequences, sequence divergence, and high computational complexity in particular when computing alignments based on RNA secondary structures. Standard alignment methods often face these issues, in particular when processing highly variable sequences or when specific phylogenetic analysis is required on selected subsequences.
We present an algorithm to determine high quality anchors that define partitions of sequences and guide the alignment of viral genomes to respect well conserved, and therefore functionally significant, regions. This new approach is implemented in the Python-based command line tool AnchoRNA, which is designed to identify conserved regions, or anchors, within coding sequences. By default, anchors are searched in translated coding sequences accounting for high mutation rates in viral genomes. AnchoRNA enhances the accuracy and efficiency of full-genome alignment by focusing on these crucial conserved regions. AnchoRNA guided alignments are systematically compared to the results of 3 alignment programs. Utilizing a dataset of 55 representative Pestivirus genomes, AnchoRNA identified 55 anchors that are used for guiding the alignment process. The incorporation of these anchors led to improvements across tested alignment tools, highlighting the effectiveness of AnchoRNA in enhancing alignment quality, especially in viral genomes.
We present an algorithm to determine high quality anchors that define partitions of sequences and guide the alignment of viral genomes to respect well conserved, and therefore functionally significant, regions. This new approach is implemented in the Python-based command line tool AnchoRNA, which is designed to identify conserved regions, or anchors, within coding sequences. By default, anchors are searched in translated coding sequences accounting for high mutation rates in viral genomes. AnchoRNA enhances the accuracy and efficiency of full-genome alignment by focusing on these crucial conserved regions. AnchoRNA guided alignments are systematically compared to the results of 3 alignment programs. Utilizing a dataset of 55 representative Pestivirus genomes, AnchoRNA identified 55 anchors that are used for guiding the alignment process. The incorporation of these anchors led to improvements across tested alignment tools, highlighting the effectiveness of AnchoRNA in enhancing alignment quality, especially in viral genomes.
2024
Ritsch, Muriel; Eulenfeld, Tom; Lamkiewicz, Kevin; Schoen, Andreas; Weber, Friedemann; Hölzer, Martin; Marz, Manja
In: Viruses, vol. 16, iss. 8, 2024, ISSN: 1999-4915.
@article{nokey_66,
title = {Endogenous Bornavirus-like Elements in Bats: Evolutionary Insights from the Conserved Riboviral L-Gene in Microbats and Its Antisense Transcription in \textit{Myotis daubentonii}},
author = {Muriel Ritsch and Tom Eulenfeld and Kevin Lamkiewicz and Andreas Schoen and Friedemann Weber and Martin Hölzer and Manja Marz},
doi = {10.3390/v16081210},
issn = {1999-4915},
year = {2024},
date = {2024-07-27},
urldate = {2024-07-27},
journal = {Viruses},
volume = {16},
issue = {8},
abstract = {Bats are ecologically diverse vertebrates characterized by their ability to host a wide range of viruses without apparent illness and the presence of numerous endogenous viral elements (EVEs). EVEs are well preserved, expressed, and may affect host biology and immunity, but their role in bat immune system evolution remains unclear. Among EVEs, endogenous bornavirus-like elements (EBLs) are bornavirus sequences integrated into animal genomes. Here, we identified a novel EBL in the microbat \textit{Myotis daubentonii}, EBLL-Cultervirus.10-MyoDau (short name is CV.10-MyoDau) that shows protein-level conservation with the L-protein of a \textit{Cultervirus} (Wuhan sharpbelly bornavirus). Surprisingly, we discovered a transcript on the antisense strand comprising three exons, which we named AMCR-MyoDau. The active transcription in \textit{Myotis daubentonii} tissues of AMCR-MyoDau, confirmed by RNA-Seq analysis and RT-PCR, highlights its potential role during viral infections. Using comparative genomics comprising 63 bat genomes, we demonstrate nucleotide-level conservation of CV.10-MyoDau and AMCR-MyoDau across various bat species and its detection in 22 \textit{Yangochiropera<i/> and 12 \textit{Yinpterochiroptera} species. To the best of our knowledge, this marks the first occurrence of a conserved EVE shared among diverse bat species, which is accompanied by a conserved antisense transcript. This highlights the need for future research to explore the role of EVEs in shaping the evolution of bat immunity.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Bats are ecologically diverse vertebrates characterized by their ability to host a wide range of viruses without apparent illness and the presence of numerous endogenous viral elements (EVEs). EVEs are well preserved, expressed, and may affect host biology and immunity, but their role in bat immune system evolution remains unclear. Among EVEs, endogenous bornavirus-like elements (EBLs) are bornavirus sequences integrated into animal genomes. Here, we identified a novel EBL in the microbat Myotis daubentonii, EBLL-Cultervirus.10-MyoDau (short name is CV.10-MyoDau) that shows protein-level conservation with the L-protein of a Cultervirus (Wuhan sharpbelly bornavirus). Surprisingly, we discovered a transcript on the antisense strand comprising three exons, which we named AMCR-MyoDau. The active transcription in Myotis daubentonii tissues of AMCR-MyoDau, confirmed by RNA-Seq analysis and RT-PCR, highlights its potential role during viral infections. Using comparative genomics comprising 63 bat genomes, we demonstrate nucleotide-level conservation of CV.10-MyoDau and AMCR-MyoDau across various bat species and its detection in 22 Yangochiropera<i/> and 12 Yinpterochiroptera species. To the best of our knowledge, this marks the first occurrence of a conserved EVE shared among diverse bat species, which is accompanied by a conserved antisense transcript. This highlights the need for future research to explore the role of EVEs in shaping the evolution of bat immunity.