IGLOO: Slicing the Features Space to Represent Sequences

Sourkov, Vsevolod

doi:10.48550/arxiv.1807.03402

Cited by 2 publications

(2 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To generate vector representations of the inputs, geNomad employs an encoder based on the IGLOO architecture 11 , which is able to extract patterns that are useful for classification from the sequence data and encode them into an embedding space (Figure 1B, Supplementary Figure 1). The IGLOO encoder begins processing one-hot-encoded matrices by applying 128 convolutional filters to generate sequence feature maps.…”

Section: Resultsmentioning

confidence: 99%

“…The IGLOO architecture has demonstrated superior performance compared to traditional alternatives (such as recurrent neural networks and convolutional neural networks) when applied to sequence data. This is attributed to its capability to gather information from non-local relationships across the entire sequence to create a global representation 11, 12 .…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

You can move, but you can't hide: identification of mobile genetic elements with geNomad

Camargo

Roux

Schulz

et al. 2023

Preprint

View full text Add to dashboard Cite

Identifying and characterizing mobile genetic elements (MGEs) in sequencing data is essential for understanding their diversity, ecology, biotechnological applications, and impact on public health. Here, we introduce geNomad, a classification and annotation framework that combines information from gene content and a deep neural network to identify sequences of plasmids and viruses. geNomad uses a large dataset of marker proteins to provide functional gene annotation and taxonomic assignment of viral genomes. Using a conditional random field model, geNomad also detects proviruses integrated into host genomes with high precision. In benchmarks that included diverse MGE and chromosome sequences, geNomad significantly outperformed other tools in all evaluated clades of plasmids and viruses. Leveraging geNomad's speed and scalability, we were able to process public metagenomes and metatranscriptomes, leading to the discovery of millions of new viruses and plasmids that are available through the IMG/VR and IMG/PR databases. We anticipate that geNomad will enable further advancements in MGE research, and it is available at https://portal.nersc.gov/genomad.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

You can move, but you can't hide: identification of mobile genetic elements with geNomad

Camargo

Roux

Schulz

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

Landscape of the metaplasmidome of deep-sea hydrothermal vents located at Arctic Mid-Ocean Ridges in the Norwegian–Greenland Sea: ecological insights from comparative analysis of plasmid identification tools

Ciuchcinski,

Stokke,

Steen

et al. 2024

FEMS Microbiology Ecology

View full text Add to dashboard Cite

Plasmids are one of the key drivers of microbial adaptation and evolution. However, their diversity and role in adaptation, especially in extreme environments, remains largely unexplored. In this study, we aimed to identify, characterize and compare plasmid sequences originating from samples collected from deep-sea hydrothermal vents located in Arctic Mid-Ocean Ridges. To achieve this, we employed, and benchmarked three recently developed plasmid identification tools - PlasX, GeNomad and PLASMe – on metagenomic data from this unique ecosystem. To date, this is the first direct comparison of these computational methods in the context of data from extreme environments. Upon recovery of plasmid contigs, we performed a multi-approach analysis, focusing on identifying taxonomic and functional biases within datasets originating from each tool. Next, we implemented a majority voting system to identify high-confidence plasmid contigs, enhancing the reliability of our findings. By analyzing the consensus plasmid sequences, we gained insights into their diversity, ecological roles, and adaptive significance. Within the high-confidence sequences, we identified a high abundance of Pseudomonadota and Campylobacterota, as well as multiple toxin-antitoxin systems. Our findings ensure a deeper understanding of how plasmids contribute to shaping microbial communities living under extreme conditions of hydrothermal vents, potentially uncovering novel adaptive mechanisms.

show abstract

IGLOO: Slicing the Features Space to Represent Sequences

Cited by 2 publications

References 12 publications

You can move, but you can't hide: identification of mobile genetic elements with geNomad

You can move, but you can't hide: identification of mobile genetic elements with geNomad

Landscape of the metaplasmidome of deep-sea hydrothermal vents located at Arctic Mid-Ocean Ridges in the Norwegian–Greenland Sea: ecological insights from comparative analysis of plasmid identification tools

Contact Info

Product

Resources

About