SemiBin, a new efficient metagene binning tool based on twin neural network

Nature Communications [IF:14. 919] ① SemiBin is a semi-supervised metagenomic binning tool based on deep learning, which supports three modes: single-sample assembly results, multi-sample co-assembly results, and multi-sample information aggregation; ② SemiBin in simulated and real datasets The binning performance is significantly better than the existing tools (such as the commonly used binning software Metabat2, Maxbin2, SolidBin and VAMB); ③ SemiBin can find differences between common Bacteroides strains from human and canine intestinal samples; ④ When using SemiBin, It is recommended to select the learning model established by the samples in the same habitat as the input samples . A deep siamese neural network improves metagenome-assembled genomes in microbiome datasets across different environments 04-28, doi: 10. 1038/s41467-022-29843-y [Editor's comment] In recent years, macros can be recovered from metagenome samples through binning techniques Genome assembled genomes (MAGs), which greatly expand the human and animal gut reference genomes . Recently, Luis Pedro Coelho and Zhao Xingming of Fudan University's Brain-inspired Intelligence Science and Technology Research Institute and their team published research in Nature Communications. They developed an efficient metagene analysis based on twin neural network. A new binning tool - SemiBin (https://github. com/BigDataBiology/SemiBin_benchmark), found that SemiBin's binning performance is significantly better than existing tools such as Metabat2, Maxbin2 and VAMB in both simulated and real datasets . In conclusion, the development of SemiBin provides new methods and insights for recovering high-quality genomes from metagenomic data .   Disclaimer: This article only represents the author's personal opinion and has nothing to do with China Probiotics Network . Its originality and the text and content stated in the text have not been verified by this site, and this site does not make any guarantee or commitment to the authenticity, completeness and timeliness of this text and all or part of its content and text. Readers are only for reference and please Verify the relevant content yourself .   Copyright Notice 1. Some articles reproduced on this site are not original, and their copyright and responsibility belong to the original author . 2. All reprinted articles, links and pictures on this website are for the purpose of conveying more information, and the source and author are clearly indicated. Media or individuals who do not wish to be reprinted can contact us for infringing information that can provide sufficient evidence. , bio149 will be deleted within 12 hours after confirmation . 3. Users are welcome to submit original articles to 86371366@qq. com, which will be published on the homepage after review, and the copyright and responsibility of the articles belong to the sender .

Nature Communications

① SemiBin is a semi-supervised metagenomic binning tool based on deep learning, which supports three modes: single-sample assembly results, multi-sample co-assembly results, and multi-sample information aggregation;

② SemiBin in simulated and real datasets The binning performance is significantly better than the existing tools (such as the commonly used binning software Metabat2, Maxbin2, SolidBin and VAMB);

③ SemiBin can find differences between common Bacteroides strains from human and canine intestinal samples;

④ When using SemiBin, It is recommended to select the learning model established by the samples in the same habitat as the input samples
.


A deep siamese neural network improves metagenome-assembled genomes in microbiome datasets across different environments
04-28, doi: 10.
1038/s41467-022-29843-y

[Editor's comment] In recent years, macros can be recovered from metagenome samples through binning techniques Genome assembled genomes (MAGs), which greatly expand the human and animal gut reference genomes
.
Recently, Luis Pedro Coelho and Zhao Xingming of Fudan University's Brain-inspired Intelligence Science and Technology Research Institute and their team published research in Nature Communications.
They developed an efficient metagene analysis based on twin neural network.
A new binning tool - SemiBin (https://github.
com/BigDataBiology/SemiBin_benchmark), found that SemiBin's binning performance is significantly better than existing tools such as Metabat2, Maxbin2 and VAMB in both simulated and real datasets
.
In conclusion, the development of SemiBin provides new methods and insights for recovering high-quality genomes from metagenomic data
.

---

Nature Communications

① SemiBin is a semi-supervised metagenomic binning tool based on deep learning, which supports three modes: single-sample assembly results, multi-sample co-assembly results, and multi-sample information aggregation;

② SemiBin in simulated and real datasets The binning performance is significantly better than the existing tools (such as the commonly used binning software Metabat2, Maxbin2, SolidBin and VAMB);

③ SemiBin can find differences between common Bacteroides strains from human and canine intestinal samples;

④ When using SemiBin, It is recommended to select the learning model established by the samples in the same habitat as the input samples
.


A deep siamese neural network improves metagenome-assembled genomes in microbiome datasets across different environments
04-28, doi: 10.
1038/s41467-022-29843-y

[Editor's comment] In recent years, macros can be recovered from metagenome samples through binning techniques Genome assembled genomes (MAGs), which greatly expand the human and animal gut reference genomes
.
Recently, Luis Pedro Coelho and Zhao Xingming of Fudan University's Brain-inspired Intelligence Science and Technology Research Institute and their team published research in Nature Communications.
They developed an efficient metagene analysis based on twin neural network.
A new binning tool - SemiBin (https://github.
com/BigDataBiology/SemiBin_benchmark), found that SemiBin's binning performance is significantly better than existing tools such as Metabat2, Maxbin2 and VAMB in both simulated and real datasets
.
In conclusion, the development of SemiBin provides new methods and insights for recovering high-quality genomes from metagenomic data
.

---

Nature Communications

---