The DOI matches its Crossref record; 31 of the 31 references it cites were found in a registry or an index.
Protein-level assembly increases protein sequence recovery from metagenomic samples manyfold
Open accessPublished version of a preprint
Abstract
The open-source de novo protein-level assembler, Plass ( https://plass.mmseqs.com ), assembles six-frame-translated sequencing reads into protein sequences. It recovers 2–10 times more protein sequences from complex metagenomes and can assemble huge datasets. We assembled two redundancy-filtered reference protein catalogs, 2 billion sequences from 640 soil samples (soil reference protein catalog) and 292 million sequences from 775 marine eukaryotic metatranscriptomes (marine eukaryotic reference catalog), the largest free collections of protein sequences. The protein-level assembler can assemble protein catalogs from raw metagenomic sequencing data, enabling large-scale metagenomics studies.
Overview
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
FLASH: fast length adjustment of short reads to improve genome assemblies
Magoč, T., & Salzberg, S. L. (2011) · Bioinformatics · doi:10.1093/bioinformatics/btr507 · Cited by 15,366
Prodigal: prokaryotic gene recognition and translation initiation site identification
Hyatt, D., Chen, G.-L., LoCascio, P. F., Land, M., Larimer, F. W., & Hauser, L. (2010) · BMC Bioinformatics · doi:10.1186/1471-2105-11-119 · Cited by 12,692
Li, D., Liu, C.-M., Luo, R., Sadakane, K., & Lam, T. (2015) · Bioinformatics · doi:10.1093/bioinformatics/btv033 · Cited by 10,133
Velvet: Algorithms for de novo short read assembly using de Bruijn graphsPreprint
Zerbino, D. R., & Birney, E. (2008) · Genome Research · doi:10.1101/gr.074492.107 · Cited by 9,587
KEGG: new perspectives on genomes, pathways, diseases and drugs
Kanehisa, M., Furumichi, M., Tanabe, M., Sato, Y., & Morishima, K. (2017) · Nucleic Acids Research · doi:10.1093/nar/gkw1092 · Cited by 8,865
MMseqs2 enables sensitive protein sequence searching for the analysis of massive data sets
Steinegger, M., & Söding, J. (2017) · Nature Biotechnology · doi:10.1038/nbt.3988 · Cited by 4,840
Self-citationmetaSPAdes: a new versatile metagenomic assembler
Nurk, S., Meleshko, D., Korobeynikov, A., & Pevzner, P. A. (2017) · Genome Research · doi:10.1101/gr.213959.116 · Cited by 4,496
Structure and function of the global ocean microbiome
Sunagawa, S., Coelho, L. P., Chaffron, S., Kultima, J. R., Labadie, K., Salazar, G., Djahanschiri, B., Zeller, G., Mende, D. R., Alberti, A., et al. (2015) · Science · doi:10.1126/science.1261359 · Cited by 2,978
Fast Genome-Wide Functional Annotation through Orthology Assignment by eggNOG-Mapper
Huerta-Cepas, J., Forslund, K., Coelho, L. P., Szklarczyk, D., Jensen, L. J., von Mering, C., & Bork, P. (2017) · Molecular Biology and Evolution · doi:10.1093/molbev/msx148 · Cited by 2,709
An integrated catalog of reference genes in the human gut microbiome
MetaHIT Consortium, Li, J., Jia, H., Cai, X., Zhong, H., Feng, Q., Sunagawa, S., Arumugam, M., Kultima, J. R., Prifti, E., et al. (2014) · Nature Biotechnology · doi:10.1038/nbt.2942 · Cited by 2,164
Cited by
Sorted by citations
Items 201–300 of 446
Strain-resolved de-novo metagenomic assembly of viral genomes and microbial 16S rRNAs
Jochheim, A., Jochheim, F. A., Kolodyazhnaya, A., Morice, É., Steinegger, M., & Söding, J. (2024) · Microbiome · doi:10.1186/s40168-024-01904-y · Cited by 13
“As in the protein-level assembler Plass [28], the alignments that satisfy the specified criteria of maximum E-value, minimum amino acid sequence identity, and minimum nucleotide sequence identity, are further considered as sequence extensions.”
MethodInfluentialSelf-citationLu, M., Wei, L., & Wu, R. (2025) · International Journal of Biological Macromolecules · doi:10.1016/j.ijbiomac.2025.140096 · Cited by 15
Comparative Assessment of Protein Large Language Models for Enzyme Commission Number Prediction
Capela, J., Zimmermann‐Kogadeeva, M., van Dijk, A. D. J., de Ridder, D., Dias, Ó., & Rocha, M. (2025) · BMC Bioinformatics · doi:10.1186/s12859-025-06081-9 · Cited by 13
Zhai, S., Zhao, H., Wang, J., Lin, S., Liu, T., Gu, S., Jiang, D., Liu, H., Kang, Y., Yao, X., et al. (2025) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.5c01084 · Cited by 21
Wang, L., Patena, W., Baalen, K. A. V., Xie, Y., Singer, E. R., Gavrilenko, S., Warren-Williams, M., Han, L., Harrigan, H. R., Chen, V. K., et al. (2022) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2022.05.31.493820 · Cited by 13
“We built our new predictor, PB-Chlamy, based on ProtBertBFD (Elnaggar et al., 2021), a natural language processing model of protein features pre-trained on the BFD database (Steinegger et al., 2019) containing 2.5 billion protein sequences from diverse organisms.”
MethodInfluentialRNA contact prediction by data efficient deep learning
Taubert, O., von der Lehr, F., Bazarova, A., Faber, C., Knechtges, P., Weiel, M., Debus, C., Coquelin, D., Basermann, A., Streit, A., et al. (2023) · Communications Biology · doi:10.1038/s42003-023-05244-9 · Cited by 11
“Since the last CASP competition for protein structure prediction 34 , the most successful methods 8,35 harness the vast amounts of sequence and structure data 36 – 38 to train large end-to-end 5 transformer 39 models, where DeepMind ’ s attention-based AF2 program is the most prominent example.”
MethodZhu, X., Liu, J., Xue, C., Tian, J., & Zhang, X. (2021) · Frontiers in Microbiology · doi:10.3389/fmicb.2020.603692 · Cited by 12
Allenby, A., Cunningham, M. R., Hillebrand‐Voiculescu, A., Comte, J., Doherty, R., & Kumaresan, D. (2022) · Frontiers in Ecology and Evolution · doi:10.3389/fevo.2022.909865 · Cited by 12
“Metagenome sequences were individually assembled by Protein-Level ASSembler (PLASS; version 4.687d7), on a protein level, using a minimum assembly length of 40 amino acids (Steinegger et al., 2018).”
MethodRevisiting evolutionary trajectories and the organization of the Pleolipoviridae family
Alarcón-Schumacher, T., Lücking, D., & Erdmann, S. (2023) · PLoS Genetics · doi:10.1371/journal.pgen.1010998 · Cited by 12
“Functional annotation of predicted viral proteins was performed InterProScan v5 [73], with DRAM [74], the package HH-suite3 [75] against the PDB70 database (release pdb70 200108) and the BFD database [76].”
Improving viral annotation with artificial intelligence
Flamholz, Z., Li, C., & Kelly, L. (2024) · mBio · doi:10.1128/mbio.03206-23 · Cited by 12
“BFD is used in AlphaFold for protein structure prediction (43), for protein-level assembly of metagenomic sequences (64), and for clustering large protein data sets (65).”
MethodZhu, L., Yang, Q., & Yang, S. (2024) · International Journal of Biological Macromolecules · doi:10.1016/j.ijbiomac.2024.136172 · Cited by 12
UniAMP: enhancing AMP prediction using deep neural networks with inferred information of peptides
Chen, Z., Ji, C., Xu, W., Gao, J., Huang, J., Xu, H., Qian, G., & Huang, J. (2025) · BMC Bioinformatics · doi:10.1186/s12859-025-06033-3 · Cited by 14
“ProtT5 was trained on a dataset of approximately 2.1 billion protein sequences, utilizing 45 million sequences from UniRef50 and 2,122 million sequences from the Big Fantastic Data-base (BFD [47]).”
InfluentialCho, J., Lee, G. M., Ko, S., Kim, Y., & Kim, D. (2025) · Journal of Virology · doi:10.1128/jvi.01901-24 · Cited by 14
Brewery: deep learning and deeper profiles for the prediction of 1D protein structure annotations
Torrisi, M., & Pollastri, G. (2020) · Bioinformatics · doi:10.1093/bioinformatics/btaa204 · Cited by 13
Cui, Z. Y., Wang, S., He, Y., Chen, Z., & Zhang, Q. (2023) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2023.3290014 · Cited by 12
“The sequence length can be regarded as the original physical property of the polypeptide [47].”
BackgroundAbBERT: Learning Antibody Humanness via Masked Language ModelingPreprint
Vashchenko, D., Nguyen, S., Gonçalves, A. R., da Silva, F. L., Petersen, B. K., Desautels, T., & Faissol, D. (2022) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2022.08.02.502236 · Cited by 13
The Dayhoff Atlas: scaling sequence diversity for improved protein generationPreprint
Yang, K., Alamdari, S., Lee, A. J., Kaymak-Loveless, K., Char, S., Brixi, G., Domingo-Enrich, C., Wang, C., Lyu, S., Fusi, N., et al. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.07.21.665991 · Cited by 13
Haas, C. M., Jasti, N., Dosey, A. M., Allen, J. D., Gillespie, R. A., McGowan, J., Leaf, E. M., Crispin, M., DeForest, C. A., Kanekiyo, M., et al. (2025) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2409566122 · Cited by 15
“Despite the large and growing Protein Data Bank [PDB; ( 33 )], this remains a significant limitation, as available structures represent only a tiny fraction of the possible space of protein structures ( 34 – 36 ).”
MucLiPred: Multi-Level Contrastive Learning for Predicting Nucleic Acid Binding Residues of Proteins
Zhang, J., Wang, R., & Wei, L. (2024) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.3c01471 · Cited by 12
Qi, D., Chen, S., & Liu, T. (2024) · Analytical Biochemistry · doi:10.1016/j.ab.2024.115603 · Cited by 11
Completeness and Consistency in Structural Domain Classifications
Schaeffer, R. D., Kinch, L. N., Pei, J., Medvedev, K. E., & Grishin, N. V. (2021) · ACS Omega · doi:10.1021/acsomega.1c00950 · Cited by 11
ACP-DRL: an anticancer peptides recognition method based on deep representation learning
Xu, X., Li, C., Yuan, X., Zhang, Q., Liu, Y., Zhu, Y., & Chen, T. (2024) · Frontiers in Genetics · doi:10.3389/fgene.2024.1376486 · Cited by 11
Russ, L., Andreo‐Jiménez, B., Nijhuis, E. H., & Postma, J. (2024) · FEMS Microbiology Ecology · doi:10.1093/femsec/fiae024 · Cited by 11
“Reads were assembled into proteins directly using the protein-level assembly tool PLASS (Steinegger et al., 2019) in paired-end mode to assemble QC-passed reads directly into proteins.”
MethodExperimental and computational approaches for membrane protein insertion and topology determination
Duart, G., Graña‐Montes, R., Pastor‐Cantizano, N., & Mingarro, I. (2024) · Methods · doi:10.1016/j.ymeth.2024.03.012 · Cited by 11
Transfer learning on protein language models improves antimicrobial peptide classification
Georgoulis, E., Zervou, M. A., & Pantazis, Y. (2025) · Scientific Reports · doi:10.1038/s41598-025-21223-y · Cited by 13
“3 is that ProtT5xl, which has been further trained with UniRef50, is significantly better from the ProtT5xl-bfd which is trained solely on the BFD dataset.”
InfluentialThafar, M. A., Albaradei, S., Uludağ, M., Alshahrani, M., Gojobori, T., Essack, M., & Gao, X. (2023) · Frontiers in Genetics · doi:10.3389/fgene.2023.1139626 · Cited by 11
“ProtBert-BFD was trained on the BFD-100 dataset (Steinegger and Söding, 2018; Steinegger et al., 2019) containing up to 2,122 million protein sequences for the protein language model (LM).”
MethodLiyaqat, T., Ahmad, T., & Saxena, C. (2023) · Journal of Computer-Aided Molecular Design · doi:10.1007/s10822-023-00533-1 · Cited by 10
“The model is pre-trained on BFD (Big Fantastic Database) dataset [44, 45] with over 2, 122 million proteins, 400 billion amino-acids and uses BERT (Bidirectional Encoder Representation from Transformers) NLP model.”
MethodThe Phenix‐AlphaFold webservice: Enabling AlphaFold predictions for use in Phenix
Poon, B. K., Terwilliger, T. C., & Adams, P. D. (2024) · Protein Science · doi:10.1002/pro.4992 · Cited by 9
“…et al., 2017), and UniRef90 (Suzek et al., 2015) clustering of the sequences in UniProt at 30% and 90% sequence identity, respectively, the MGnify sequence database (Richardson et al., 2023), and the BFD sequence database (Jumper et al., 2021; Steinegger et al., 2019; Steinegger & Söding, 2018).”
MethodDohan, D., Gane, A., Bileschi, M. L., Belanger, D., & Colwell, L. J. (2021) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3447548.3467163 · Cited by 13
“Relatedly, Elnaggar et al. [19] compared models trained on UniRef100 and BFD, and found that training on the larger dataset (2.12B sequences in BFD vs. 216M sequences in UniRef100) improved downstream task performance, but not significantly.”
MethodWoodard, J. C., Zheng, W. H., & Zhang, Y. (2021) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1009370 · Cited by 10
“PSSM generation To construct the PSSM, multiple sequence alignments (MSAs) for the query sequence were first generated by DeepMSA2, which utilizes HHblits (43, 44), Jackhmmer (45) and HMMsearch (46) to iteratively search two whole-genome sequence databases (Uniclust30 (47) and UniRef90 (48)) and four metagenome sequen…”
MethodKahle, M., Appelgren, S., Elofsson, A., Carroni, M., & Ädelroth, P. (2023) · BMC Biology · doi:10.1186/s12915-023-01546-w · Cited by 10
Discovery of highly active kynureninases for cancer immunotherapy through protein language model
Eom, H., Park, S., Cho, K. S., Lee, J., Kim, H., Kim, S., Yang, J., Han, Y.-H., Lee, J., Seok, C., et al. (2025) · Nucleic Acids Research · doi:10.1093/nar/gkae1245 · Cited by 10
“1) in a database of cultivated ( 16 ) and metagenomic protein sequences from BFD ( 17 ) and Meta-clust2 ( 18 ), a total of 3 144 354 589 protein sequences, using MMseqs2’s (version 3fa46) iterative search (–num-iterations 3 -s 7.5 -c 0.9).”
InfluentialSelf-citationde Araújo, L. P., Dias, M. E. C., Scodeler, G. C., Santos, A. D. S., Soares, L. M., Corsetti, P. P., Padovan, A. C. B., da Silveira, N. J. F., & de Almeida, L. A. (2022) · ImmunoInformatics · doi:10.1016/j.immuno.2022.100015 · Cited by 9
“Multiple databases as UniRef90 [34], MGnify [35], BFD [36], Uniclust30 [37], PDB70 and PDB [38] were used to homology modeling the structural proteins.”
MethodExploration of protein sequence embeddings for protein-ligand binding site detection
Hoksza, D., & Gamouh, H. (2022) · IEEE International Conference on Bioinformatics and Biomedicine (BIBM) · doi:10.1109/bibm55620.2022.9995025 · Cited by 8
“…(BB) [33] that was trained on SCOP [34] and is based on a bidirectional LSTM network that blends protein sequence and structure information to learn the embeddings, ii) ProtBert that was trained on BFD [35] and iii) ProtT5-XL-UniRef50 (ProtT5) that was trained on BFD and finetuned on UniRef50.”
MethodBioinformatics-aided Protein Sequence Analysis and Engineering
Zhang, W., & Wang, T. (2023) · Current Protein and Peptide Science · doi:10.2174/1389203724666230509124300 · Cited by 9
Alg-MFDL: A multi-feature deep learning framework for allergenic proteins prediction
Hu, X., Li, J., & Liu, T. (2024) · Analytical Biochemistry · doi:10.1016/j.ab.2024.115701 · Cited by 9
IDP-Bert: Predicting Properties of Intrinsically Disordered Proteins Using Large Language Models
Mollaei, P., Sadasivam, D., Guntuboina, C., & Farimani, A. B. (2024) · The Journal of Physical Chemistry B · doi:10.1021/acs.jpcb.4c02507 · Cited by 13
Hyena architecture enables fast and efficient protein language modeling
Zhang, Y., Bian, B., & Okumura, M. (2024) · iMetaOmics. · doi:10.1002/imo2.45 · Cited by 10
“ProtBert ‐ U100 is trained on UniRef100 [52], which contains 216 million protein sequences, while ProtBert ‐ BFD is trained on BFD [53], which includes 2,122 million protein sequences.”
Xie, P., Yao, L., Guan, J., Chung, C., Zhao, Z., Long, F., Sun, Z., Lee, T.-Y., & Chiang, Y. (2025) · Protein Science · doi:10.1002/pro.70087 · Cited by 10
“On the other hand, the BFD data-base comprises approximately 2.1 billion protein sequences sourced from diverse databases, including Uniprot/TrEMBL + Swissprot (UniProt Consortium, 2023), Metaclust (Steinegger & Söding, 2018), and the Soil and Marine Eukaryotic Reference Catalogs (Steinegger et al., 2019).”
InfluentialNguyen, N. D. H., Pham, N. T., Tran, D. T., Wei, L., Malik, A., & Manavalan, B. (2025) · Journal of Cheminformatics · doi:10.1186/s13321-025-01078-1 · Cited by 11
“ProtTrans: This research focuses on training two auto-regressive models and four auto-encoder models using two key datasets, UniRef [24] and BFD (Big Fantastic Database) [25, 26].”
Liu, Z., Pan, W., Li, W., Zhen, X., Liang, J., Cai, W., Xu, F., Yuan, K., & Lin, G. N. (2022) · Biology · doi:10.3390/biology11101454 · Cited by 9
“However, due to the limited number of accurately measured protein crystal structures [12,13], most sequence-based predictors can only be trained using evolutionary conservatism profiles and predicted secondary structure.”
Kandathil, S. M., Greener, J. G., Lau, A. M., & Jones, D. T. (2020) · doi:10.1101/2020.11.27.401232 · Cited by 8
Multipurpose On‐the‐Spot Peptide‐Based Hydrogels for Skin, Cornea, and Heart Repair
Ross, A., Guo, X., Salazar, G. A. M., Schejtman, S. D. G., Hage, J. E., Comtois‐Bona, M., MacAdam, A. J., Guzmán‐Soto, I., Takaya, H., Hu, K., et al. (2024) · Advanced Functional Materials · doi:10.1002/adfm.202402564 · Cited by 8
Zhang, M., Wang, X., Xu, S., Ge, F., Paixao, I. C., Song, J., & Yu, D. (2024) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.4c00739 · Cited by 8
Paranou, D., Chatzigoulas, A., & Cournia, Z. (2024) · Bioinformatics Advances · doi:10.1093/bioadv/vbae078 · Cited by 8
“These pLMs have been trained using sequence databases such as UniRef50 (Suzek et al. 2015) and BFD (Steinegger et al. 2019), which include millions of protein sequences and billions of amino acids.”
Benchmarking HelixFold3-Predicted Holo Structures for Relative Free Energy Perturbation Calculations
Furui, K., & Ohue, M. (2025) · ACS Omega · doi:10.1021/acsomega.4c11413 · Cited by 8
Jang, Y. E., Kwon, M., Kim, S. G., George, N. P., Hwang, J. S., Basith, S., & Lee, G. (2025) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2025.110590 · Cited by 8
SeaMoon: From protein language models to continuous structural heterogeneity
Lombard, V., Timsit, D., Grudinin, S., & Laine, E. (2025) · Structure · doi:10.1016/j.str.2025.06.010 · Cited by 8
Clustering the protein universe of life using DIAMOND DeepClust
Buchfink, B., Barbe, E., Ashkenazy, H., Reuter, K., Kennedy, J. A., & Drost, H. (2026) · Nature Methods · doi:10.1038/s41592-026-03030-z · Cited by 33
Wang, X., Zhang, Z., & Liu, C. (2024) · Journal of Molecular Biology · doi:10.1016/j.jmb.2024.168810 · Cited by 8
Development of sex‐linked SSR marker in the genus Phoenix and validation in P. dactylifera
Wang, Y., Ihase, L. O., Htwe, Y. M., Shi, P., Zhang, D., Li, D., Emoghene, B. O., & Iserhienrhien, A. (2020) · Crop Science · doi:10.1002/csc2.20187 · Cited by 8
S-Pred: protein structural property prediction using MSA transformer
Hong, Y., Song, J., Ko, J., Lee, J., & Shin, W. (2022) · Scientific Reports · doi:10.1038/s41598-022-18205-9 · Cited by 7
Schön, M. E., Eme, L., & Ettema, T. J. G. (2019) · Bioinformatics · doi:10.1093/bioinformatics/btz799 · Cited by 7
Bryant, P., & Noé, F. (2023) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2023.04.15.536993 · Cited by 7
Kagaya, Y., Nakamura, T., Verburgt, J. C., Jain, A., Terashi, G., Punuru, P., Tugolukova, E., Park, J. H., Saha, A., Huang, D. Y. C., et al. (2025) · Proteins Structure Function and Bioinformatics · doi:10.1002/prot.70033 · Cited by 8
Chen, L., Xun, X., & Zhou, B. (2026) · Scientific Reports · doi:10.1038/s41598-026-35110-7 · Cited by 7
“ProtT5 is a 24-layer transformer-based language model that was initially pre-trained on a comprehensive protein dataset from the Big Fantastic Database (BFD) 19,20 , and subsequently fine-tuned using the UniRef 50 dataset 21 .”
Wang, B., & Li, W. (2024) · Genes · doi:10.3390/genes15081090 · Cited by 7
“Additionally, other methods like ProtT5-XL-U50 have been pre-trained on datasets like BFD [102], yielding 1024-dimensional sequence embeddings normalized to scores between 0 and 1.”
MethodFALCON2: a web server for high-quality prediction of protein tertiary structures
Kong, L., Ju, F., Zhang, H., Sun, S., & Bu, D. (2021) · BMC Bioinformatics · doi:10.1186/s12859-021-04353-8 · Cited by 8
Suviriyapaisal, N., & Wichadakul, D. (2023) · RSC Advances · doi:10.1039/d3ra03796g · Cited by 8
Petabase-Scale Homology Search for Structure PredictionPreprint
Lee, S., Kim, G., Karin, E. L., Mirdita, M., Park, S., Chikhi, R., Babaian, A., Kryshtafovych, A., & Steinegger, M. (2024) · Cold Spring Harbor Perspectives in Biology · doi:10.1101/cshperspect.a041465 · Cited by 9
Self-citationEvaluating variant effect prediction across virusesPreprint
Gurev, S. F., Youssef, N., Jain, N., Mehrotra, A., Leung, S. R. M., Jackson, A., & Marks, D. S. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.08.04.668549 · Cited by 10
Genome-scale annotation of protein binding sites via language model and geometric deep learning
Yuan, Q., Tian, C., & Yang, Y. (2024) · eLife · doi:10.7554/elife.93695.3 · Cited by 9
Decoding the effects of mutation on protein interactions using machine learning
Wang, X., Li, A., Zhao, Y., & Peng, Y. (2025) · Biophysics Reviews · doi:10.1063/5.0249920 · Cited by 7
Zhang, M., Yang, Q., Lou, J., Hu, Y., & Shi, Y. (2025) · Biochimica et Biophysica Acta (BBA) - Molecular Basis of Disease · doi:10.1016/j.bbadis.2025.167772 · Cited by 6
Guan, M., Han, J., Zhang, S., Zheng, H., & Liu, J. (2025) · Research · doi:10.34133/research.0773 · Cited by 6
Protein foundation models: a comprehensive survey
Xu, H., Li, L., Pan, S., Cheng, P., Wang, Y., Rong, Z., Feng, L., Huang, X., Wang, S., & Shu, W. (2026) · Science China Life Sciences · doi:10.1007/s11427-025-3147-2 · Cited by 8
Exploring the Knowledge of an Outstanding Protein to Protein Interaction Transformer
Yang, S., Cheng, P., Liu, Y., Feng, D., & Wang, S. (2024) · IEEE/ACM Transactions on Computational Biology and Bioinformatics · doi:10.1109/tcbb.2024.3381825 · Cited by 6
“The model containing 3 billion parameters is pre-trained on the BFD [40] dataset and fine-tuned on the UniRef50 [41] dataset which contains 45 and 2122 million proteins, respectively.”
MethodLiang, X., Zhao, H., & Wang, J. (2024) · Lecture notes in computer science · doi:10.1007/978-981-97-5131-0_6 · Cited by 6
Exploring Evolution to Uncover Insights Into Protein Mutational Stability
Hermans, P., Tsishyn, M., Schwersensky, M., Rooman, M., & Pucci, F. (2025) · Molecular Biology and Evolution · doi:10.1093/molbev/msae267 · Cited by 6
“We primarily used the UniRef90 set (Suzek et al. 2015), but also tested searching in two larger datasets: UniRef100 and Metagenomics, a metagenomic dataset constructed by merging UniRef90 (Suzek et al. 2015), the Big Fantastic Database (Steinegger et al. 2019), the Gut Phage Database (Camarillo-Guerrero et al. 2021),…”
Zhang, L., & Liu, T. (2025) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.5c00016 · Cited by 7
Liu, J., He, G.-X., Zhao, K., & Zhang, G. (2022) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2022.01.11.475831 · Cited by 7
Molecular Display of the Animal Meta-Venome for Discovery of Novel Therapeutic Peptides
Hsiao, M.-H., Miao, Y., Liu, Z., Schütze, K., Limjunyawong, N., Chien, D. C.-C., Monteiro, W. D., Chu, L.-S., Morgenlander, W., Jayaraman, S., et al. (2025) · Molecular & Cellular Proteomics · doi:10.1016/j.mcpro.2024.100901 · Cited by 5
Self-citationApusomonad rhodopsins: A new family of ultraviolet to blue light–absorbing rhodopsin channels
Galindo, L. J., Takaramoto, S., Nagata, T., Rozenberg, A., Takahashi, H., Béjà, O., & Inoue, K. (2025) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2510619122 · Cited by 6
“…data from the Tara Oceans metagenomes and metatranscriptomes (48, 49): one ApuR type closely related to Tt ApuR from Thecamonas trahens and a distinct type found predominantly at station TARA_39 in the North Indian Ocean [represented in our dataset by Tara ApuR derived from the MERC assembly (50)].”
Liang, Y., & Wang, C. (2026) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.5c02745 · Cited by 5
Arsenic oxidation by root endophytes mediates arsenic speciation within rice (Oryza sativa)
Wang, Z., Song, B., Sun, X., Huang, D., Häggblom, M. M., Yu, Z., Kong, T., He, B., He, B., Zhang, H., et al. (2026) · Environmental Pollution · doi:10.1016/j.envpol.2026.128194 · Cited by 5
Qi, D., & Liu, T. (2024) · Biochimica et Biophysica Acta (BBA) - General Subjects · doi:10.1016/j.bbagen.2024.130721 · Cited by 5
Screening great ape museum specimens for DNA viruses
Hämmerle, M., Guellil, M., Trgovec-Greif, L., Cheronet, O., Sawyer, S., Ruiz‐Gartzia, I., Lizano, E., Rymbekova, A., Gelabert, P., Bernardi, P., et al. (2024) · Scientific Reports · doi:10.1038/s41598-024-80780-w · Cited by 6
Le, V., Malik, M. S., Lin, Y.-J., Liu, Y., Chang, Y., & Ou, Y. (2024) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2024.109541 · Cited by 5
Lv, Z., Wei, M., Pei, H., Peng, S., Li, M., & Jiang, L. (2024) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2024.109598 · Cited by 5
Language model-accelerated deep symbolic optimization
da Silva, F. L., Gonçalves, A. R., Nguyen, S., Vashchenko, D., Glatt, R., Desautels, T., Landajuela, M., Faissol, D., & Petersen, B. K. (2023) · Neural Computing and Applications · doi:10.1007/s00521-023-08802-8 · Cited by 5
“Prot-BERT was trained on UniRef100 [33] and BFD-100 [34] datasets which consist of more than 1 billion of protein sequences.”
Li, X., Wang, G.-A., Wei, Z., Wang, H., & Zhu, X. (2023) · Computational Biology and Chemistry · doi:10.1016/j.compbiolchem.2023.107970 · Cited by 5
Gao, M., Song, C., & Liu, T. (2024) · Journal of Cellular Biochemistry · doi:10.1002/jcb.30642 · Cited by 5
BEAN and HABAS : Polyphyletic insertions in the DNA ‐directed RNA polymerase
Álvarez-Carreño, C., Huynh, A. T., Petrov, A. S., Orengo, C. A., & Williams, L. D. (2024) · Protein Science · doi:10.1002/pro.5194 · Cited by 4
The COMER web server for protein analysis by homology
Dapkūnas, J., & Margelevičius, M. (2022) · Bioinformatics · doi:10.1093/bioinformatics/btac807 · Cited by 5
MSAGPT: Neural Prompting Protein Structure Prediction via MSA Generative Pre-TrainingPreprint
Chen, B., Bei, Z., Cheng, X., Li, P., Tang, J., & Song, L. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.06.10.598380 · Cited by 11
“The success of these cutting-edge methods largely rely on the utilization of MSA, which are retrieved through search algorithms [24, 25, 26, 27] against vast databases [13, 14, 15, 16].”
Controllable All-Atom Protein Generation with Latent DiffusionPreprint
Lu, A. X., Yang, J., Long, Y., Lambert, T., Yan, W., Robinson, S. A., Yang, K., Gligorijević, V., Cho, K., Bonneau, R., et al. (2024) · Research Square · doi:10.1101/2024.12.02.626353 · Cited by 6
Li, Y., Nan, X., Zhang, S., Zhou, Q., Lu, S., & Tian, Z. (2024) · Biomolecules · doi:10.3390/biom14101220 · Cited by 4
“ProtT5 is pre-trained on the BFD dataset [83] and subsequently fine-tuned on UniRef50 [84].”
MethodFlexible protein–ligand docking with diffusion-based side-chain packing
Zhang, R., Jiang, X., Cao, D., Wang, Z., Yu, J., Chen, M., Fan, Z., Kong, X., Xiong, J., Zhang, Z., et al. (2025) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2511925122 · Cited by 4
“Over the past decades, experimental methods have resolved approximately 100,000 protein structures (64), representing only a small fraction of the known protein sequence space (65).”
Peptide cheminformatics tools: making computational tasks accessible in peptide drug discovery
Erckes, V., Abderrahmane, M., Jusot, M., Steuer, C., & Ochoa, R. (2026) · Drug Discovery Today · doi:10.1016/j.drudis.2026.104612 · Cited by 6
Yu, X., Ge, F., Yu, D., & Deng, Z. (2026) · Briefings in Bioinformatics · doi:10.1093/bib/bbag050 · Cited by 4
“ProtTrans [30], which was developed by using the UniRef and Big Fantastic Database dataset with up to 393 billion amino acids [33, 34], leverages 1024-dimensional embeddings per residue.”
Bochtler, M. (2024) · BioEssays · doi:10.1002/bies.202400155 · Cited by 6
GraphPhos: Predict Protein-Phosphorylation Sites Based on Graph Neural Networks
Wang, Z., Yang, X., Gao, S., Liang, Y., & Shi, X. (2025) · International Journal of Molecular Sciences · doi:10.3390/ijms26030941 · Cited by 4
“The model had been trained on two protein data sets: UniRef100 [49] and BFD100 [50], respectively.”
Revolutionized virome research using systems microbiology approaches
Chitcharoen, S., Sivapornnukul, P., & Payungporn, S. (2022) · Experimental Biology and Medicine · doi:10.1177/15353702221102895 · Cited by 4
Protein Subcellular Localization Prediction by Combining ProtBert and BiGRU
Wang, Z., Lin, T., Yang, X., Liang, Y., & Shi, X. (2022) · IEEE International Conference on Bioinformatics and Biomedicine (BIBM) · doi:10.1109/bibm55620.2022.9995180 · Cited by 5
“proposed a pre-trained protein representation model - ProtBert [8] which is obtained via finetuning the original Bert [9]model with the protein sequences from UniRef100 [10] and BFD [11] databases.”
MethodAnselmi, N. K., Vanyo, S. T., Clark, N. D., Rodriguez, D. M. L., Jones, M. M., Rosenthal, S., Patel, D. T., Marconi, R. T., & Visser, M. B. (2024) · Molecular Oral Microbiology · doi:10.1111/omi.12484 · Cited by 4
TMVisDB: Annotation and 3D-visualization of Transmembrane Proteins
Olenyi, T., Marquet, C., Grekova, A., Houri, L., Heinzinger, M., Dallago, C., & Rost, B. (2025) · Journal of Molecular Biology · doi:10.1016/j.jmb.2025.168997 · Cited by 4
Vajjala, M., Johnson, B., Kasparek, L., Leuze, M. R., & Yao, Q. (2022) · Frontiers in Genetics · doi:10.3389/fgene.2022.935351 · Cited by 4
“The protein sequences generated from the PLASS example dataset are duplicated to test the effects of truncated sequences.”
MethodInfluentialJung, J., Bugenyi, A. W., Lee, M.-R., Choi, Y., Song, K., Lee, H., Son, Y., Lee, D., Lee, S., Son, Y.-J., et al. (2022) · Scientific Reports · doi:10.1038/s41598-022-18503-2 · Cited by 4
Yakoubi, S. (2024) · IUBMB Life · doi:10.1002/iub.2819 · Cited by 4
“…transformer model trained on about 27 million sequences from UniRef 50, 59 and (ii) ProtT5, 60 a sequence-to-sequence model derived from the T5 architecture, 61 trained on the large Big Fan-tastic Database (BFD) 62 comprising 2.1 billion sequences and fine-tuned on the UniRef50 database.”
InfluentialSingh, J., Litfin, T., Singh, J., Paliwal, K. K., & Zhou, Y. (2021) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2021.06.19.449089 · Cited by 4
Metadata
- Authors
- Steinegger, M., Mirdita, M., & Söding, J.
- Affiliations
- Seoul National University (KR)
- Max Planck Institute for Biophysical Chemistry (DE)
- Corresponding author
- Steinegger, Martin
- Söding, Johannes
- Venue
- Nature Methods
- Published
- 2019-06-24
- Volume
- 16
- Issue
- 7
- Pages
- 603-606
- ISSN
- 1548-7091 (ISSN-L)
- 1548-7105
- OpenAlex
- W2949342052
- Cited by
- 445
- References
- 31
- Citations by year
- 2026: 38 · 2025: 78 · 2024: 101 · 2023: 78 · 2022: 71 · 2021: 50 · 2020: 23 · 2019: 6
- FWCI
- 12.42
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Locations
- Nature Methods (publishedVersion, cc-by)
- bioRxiv (Cold Spring Harbor Laboratory) (acceptedVersion, cc-by-nc-nd)
- PubMed (publishedVersion)
- MPG.PuRe (Max Planck Society) (submittedVersion)
- Seoul National University Open Repository (Seoul National University) (submittedVersion)
- Funders
- Materials and Energy Research Center
- Topics
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Genomics and Phylogenetic Studies
- Physical Sciences › Environmental Science › Ecology › Environmental DNA in Biodiversity Studies
- Physical Sciences › Environmental Science › Ecology › Microbial Community Ecology and Physiology
- Keywords
- metagenomics, metatranscriptomics, soil metagenome
- MeSH terms
- Amino Acid Sequence, Codon, Proteins, Open Reading Frames, Metagenomics
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
