Adversarial and variational autoencoders improve metagenomic binning

Research output: Contribution to journal › Journal article › Research › peer-review

Standard

Adversarial and variational autoencoders improve metagenomic binning. / Líndez, Pau Piera; Johansen, Joachim; Kutuzova, Svetlana; Sigurdsson, Arnor Ingi; Nissen, Jakob Nybo; Rasmussen, Simon.

In: Communications Biology , Vol. 6, 1073, 2023.

Research output: Contribution to journal › Journal article › Research › peer-review

Harvard

Líndez, PP, Johansen, J, Kutuzova, S, Sigurdsson, AI, Nissen, JN & Rasmussen, S 2023, 'Adversarial and variational autoencoders improve metagenomic binning', Communications Biology , vol. 6, 1073. https://doi.org/10.1038/s42003-023-05452-3

APA

Líndez, P. P., Johansen, J., Kutuzova, S., Sigurdsson, A. I., Nissen, J. N., & Rasmussen, S. (2023). Adversarial and variational autoencoders improve metagenomic binning. Communications Biology , 6, [1073]. https://doi.org/10.1038/s42003-023-05452-3

Vancouver

Líndez PP, Johansen J, Kutuzova S, Sigurdsson AI, Nissen JN, Rasmussen S. Adversarial and variational autoencoders improve metagenomic binning. Communications Biology . 2023;6. 1073. https://doi.org/10.1038/s42003-023-05452-3

Author

Líndez, Pau Piera ; Johansen, Joachim ; Kutuzova, Svetlana ; Sigurdsson, Arnor Ingi ; Nissen, Jakob Nybo ; Rasmussen, Simon. / Adversarial and variational autoencoders improve metagenomic binning. In: Communications Biology . 2023 ; Vol. 6.

Bibtex

@article{d553b5d757e64e8e9471b73f89c40798,
title = "Adversarial and variational autoencoders improve metagenomic binning",
abstract = "Assembly of reads from metagenomic samples is a hard problem, often resulting in highly fragmented genome assemblies. Metagenomic binning allows us to reconstruct genomes by re-grouping the sequences by their organism of origin, thus representing a crucial processing step when exploring the biological diversity of metagenomic samples. Here we present Adversarial Autoencoders for Metagenomics Binning (AAMB), an ensemble deep learning approach that integrates sequence co-abundances and tetranucleotide frequencies into a common denoised space that enables precise clustering of sequences into microbial genomes. When benchmarked, AAMB presented similar or better results compared with the state-of-the-art reference-free binner VAMB, reconstructing ~7% more near-complete (NC) genomes across simulated and real data. In addition, genomes reconstructed using AAMB had higher completeness and greater taxonomic diversity compared with VAMB. Finally, we implemented a pipeline Integrating VAMB and AAMB that enabled improved binning, recovering 20% and 29% more simulated and real NC genomes, respectively, compared to VAMB, with moderate additional runtime.",
author = "L{\'i}ndez, {Pau Piera} and Joachim Johansen and Svetlana Kutuzova and Sigurdsson, {Arnor Ingi} and Nissen, {Jakob Nybo} and Simon Rasmussen",
note = "Publisher Copyright: {\textcopyright} 2023, Springer Nature Limited.",
year = "2023",
doi = "10.1038/s42003-023-05452-3",
language = "English",
volume = "6",
journal = "Communications Biology",
issn = "2399-3642",
publisher = "nature publishing group",

}

RIS

TY - JOUR

T1 - Adversarial and variational autoencoders improve metagenomic binning

AU - Líndez, Pau Piera

AU - Johansen, Joachim

AU - Kutuzova, Svetlana

AU - Sigurdsson, Arnor Ingi

AU - Nissen, Jakob Nybo

AU - Rasmussen, Simon

N1 - Publisher Copyright: © 2023, Springer Nature Limited.

PY - 2023

Y1 - 2023

N2 - Assembly of reads from metagenomic samples is a hard problem, often resulting in highly fragmented genome assemblies. Metagenomic binning allows us to reconstruct genomes by re-grouping the sequences by their organism of origin, thus representing a crucial processing step when exploring the biological diversity of metagenomic samples. Here we present Adversarial Autoencoders for Metagenomics Binning (AAMB), an ensemble deep learning approach that integrates sequence co-abundances and tetranucleotide frequencies into a common denoised space that enables precise clustering of sequences into microbial genomes. When benchmarked, AAMB presented similar or better results compared with the state-of-the-art reference-free binner VAMB, reconstructing ~7% more near-complete (NC) genomes across simulated and real data. In addition, genomes reconstructed using AAMB had higher completeness and greater taxonomic diversity compared with VAMB. Finally, we implemented a pipeline Integrating VAMB and AAMB that enabled improved binning, recovering 20% and 29% more simulated and real NC genomes, respectively, compared to VAMB, with moderate additional runtime.

AB - Assembly of reads from metagenomic samples is a hard problem, often resulting in highly fragmented genome assemblies. Metagenomic binning allows us to reconstruct genomes by re-grouping the sequences by their organism of origin, thus representing a crucial processing step when exploring the biological diversity of metagenomic samples. Here we present Adversarial Autoencoders for Metagenomics Binning (AAMB), an ensemble deep learning approach that integrates sequence co-abundances and tetranucleotide frequencies into a common denoised space that enables precise clustering of sequences into microbial genomes. When benchmarked, AAMB presented similar or better results compared with the state-of-the-art reference-free binner VAMB, reconstructing ~7% more near-complete (NC) genomes across simulated and real data. In addition, genomes reconstructed using AAMB had higher completeness and greater taxonomic diversity compared with VAMB. Finally, we implemented a pipeline Integrating VAMB and AAMB that enabled improved binning, recovering 20% and 29% more simulated and real NC genomes, respectively, compared to VAMB, with moderate additional runtime.

U2 - 10.1038/s42003-023-05452-3

DO - 10.1038/s42003-023-05452-3

M3 - Journal article

C2 - 37865678

AN - SCOPUS:85174562894

VL - 6

JO - Communications Biology

JF - Communications Biology

SN - 2399-3642

M1 - 1073

ER -

ID: 371694099