PlasmidHostFinder: Prediction of Plasmid Hosts Using Random Forest

Aytan-Aktug, Derya; Clausen, Philip Thomas Lanken Conradsen; Szarvas, Judit; Munk, Patrick; Otani, Saria; Nguyen, M.; Davis, James J.; Lund, Ole; Aarestrup, Frank Møller

doi:10.1128/msystems.01180-21

Cited by 14 publications

(12 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We compared HOTSPOT with three state-of-the-art tools designed for plasmid host prediction: MOB-typer ( Robertson et al 2020 ), PlasmidHostFinder ( Aytan-Aktug et al 2022 ), and PlasFlow ( Krawczyk et al 2018 ), all of which are briefly described in Section 1.1. PlasFlow can only output phylum-level results.…”

Section: Resultsmentioning

confidence: 99%

“…They mainly adopt k -mers, GC-content, and codon usage as features. For example, PlasmidHostFinder ( Aytan-Aktug et al 2022 ) utilizes plasmid genomic signatures, such as 8-mer counts and codon frequencies, to train four random forest models corresponding to the order, family, genus, and species level. PlasFlow ( Krawczyk et al 2018 ) trained neural networks using normalized k -mer frequency vector to predict phylum-level plasmid hosts.…”

Section: Introductionmentioning

confidence: 99%

“…In each classifier, we take all computationally predicted proteins (CPs), annotated MOB/MPF proteins, and identified incompatibility (Inc) group (by replicon typing) of the query plasmid contig as input features for taxon classification. We compared our tool with the state-of-the-art approaches designed for plasmid host prediction: MOB-typer ( Robertson et al 2020 ), PlasmidHostFinder ( Aytan-Aktug et al 2022 ), and PlasFlow ( Krawczyk et al 2018 ). We rigorously tested our tool on multiple datasets, including RefSeq complete plasmids, artificial contigs, mock metagenomic data, the Hi-C dataset, and the CAMI2 marine dataset.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

HOTSPOT: hierarchical host prediction for assembled plasmid contigs with transformer

Shang

Tang

et al. 2023

Bioinformatics

View full text Add to dashboard Cite

Motivation As prevalent extrachromosomal replicons in many bacteria, plasmids play an essential role in their hosts’ evolution and adaptation. The host range of a plasmid refers to the taxonomic range of bacteria in which it can replicate and thrive. Understanding host ranges of plasmids sheds light on studying the roles of plasmids in bacterial evolution and adaptation. Metagenomic sequencing has become a major means to obtain new plasmids and derive their hosts. However, host prediction for assembled plasmid contigs still needs to tackle several challenges: different sequence compositions and copy numbers between plasmids and the hosts, high diversity in plasmids, and limited plasmid annotations. Existing tools have not yet achieved an ideal tradeoff between sensitivity and precision on metagenomic assembled contigs. Results In this work, we construct a hierarchical classification tool named HOTSPOT, whose backbone is a phylogenetic tree of the bacterial hosts from phylum to species. By incorporating the state-of-the-art language model, Transformer, in each node’s taxon classifier, the top-down tree search achieves an accurate host taxonomy prediction for the input plasmid contigs. We rigorously tested HOTSPOT on multiple datasets, including RefSeq complete plasmids, artificial contigs, simulated metagenomic data, mock metagenomic data, the Hi-C dataset, and the CAMI2 marine dataset. All experiments show that HOTSPOT outperforms other popular methods. Availability The source code of HOTSPOT is available via: https://github.com/Orin-beep/HOTSPOT Supplementary information Supplementary data are available at Bioinformatics online.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

HOTSPOT: hierarchical host prediction for assembled plasmid contigs with transformer

Shang

Tang

et al. 2023

Bioinformatics

View full text Add to dashboard Cite

show abstract

“…Plasmid hosts, on the other hand, due to the competence of plasmids (including self-replication, etc. ), are difficult to identify ( Suzuki et al, 2010 ) or can only be classified against existing databases ( Aytan-Aktug et al, 2022 ).…”

Section: Methodsmentioning

confidence: 99%

Mobilome-driven segregation of the resistome in biological wastewater treatment

Nies

Busi

Kunath

et al. 2022

eLife

View full text Add to dashboard Cite

Biological wastewater treatment plants (BWWTP) are considered to be hotspots for the evolution and subsequent spread of antimicrobial resistance (AMR). Mobile genetic elements (MGEs) promote the mobilization and dissemination of antimicrobial resistance genes (ARGs) and are thereby critical mediators of AMR within the BWWTP microbial community. At present, it is unclear whether specific AMR categories are differentially disseminated via bacteriophages (phages) or plasmids. To understand the segregation of AMR in relation to MGEs, we analyzed meta-omic (metagenomic, metatranscriptomic and metaproteomic) data systematically collected over 1.5 years from a BWWTP. Our results showed a core group of 15 AMR categories which were found across all timepoints. Some of these AMR categories were disseminated exclusively (bacitracin) or primarily (aminoglycoside, MLS and sulfonamide) via plasmids or phages (fosfomycin and peptide), whereas others were disseminated equally by both. Combined and timepoint-specific analyses of gene, transcript and protein abundances further demonstrated that aminoglycoside, bacitracin and sulfonamide resistance genes were expressed more by plasmids, in contrast to fosfomycin and peptide AMR expression by phages, thereby validating our genomic findings. In the analyzed communities, the dominant taxon Candidatus Microthrix parvicella was a major contributor to several AMR categories whereby its plasmids primarily mediated aminoglycoside resistance. Importantly, we also found AMR associated with ESKAPEE pathogens within the BWWTP, and here MGEs also contributed differentially to the dissemination of the corresponding ARGs. Collectively our findings pave the way toward understanding the segmentation of AMR within MGEs, thereby shedding new light on resistome populations and their mediators, essential elements that are of immediate relevance to human health.

show abstract

“…However, these tools can perform poorly with contigs assembled from metagenomes because they are not designed for identifying incomplete sequences ( 4 , 12 – 15 ). On the other hand, machine-learning-based approaches can be designed for identifying extrachromosomal elements using short sequence-based matching, or by identifying sequence characteristics such as GC content, codon usage, and gene density ( 12 , 14 , 16 , 17 ). Machine-learning models developed for both WGS and metagenomic reads use various algorithms such as support vector machines ( 18 ), logistic regressions ( 12 ), random forests ( 19 ), and artificial neural networks ( 14 , 15 , 20 – 23 ).…”

Section: Introductionmentioning

confidence: 99%