Peter J. Vikesland

ARGfore forecasts antibiotic resistance gene abundances from time-series metagenomic data, outperforming standard methods with lower error and better efficiency.

The paper introduces a pipeline using PPINs and machine learning to identify ARGs, showing improved discrimination of ARGs from nonARGs and their association with mobile genetic elements.

ProtAlign-ARG improves antibiotic resistance gene detection by combining protein language models with alignment-based scoring, achieving higher recall than existing methods.

MetaCompare 2.0 differentiates human health and ecological resistome risks using metagenomic data, showing higher risk scores for human-contaminated environments.

CIWARS is a web server for analyzing longitudinal metagenomic data to detect antibiotic resistance genes and assess AR risks in environmental samples.

Kairos detects in situ horizontal gene transfer in microbiomes using time-series metagenomic data and competitive read alignment, improving detection of antibiotic resistance genes compared to MAG-based methods.

The paper introduces mobileOG-db, a manually curated database of protein families involved in bacterial mobile genetic element life cycles, enabling improved MGE annotation and analysis.

MetaMLP is a machine learning classifier for metagenomic functional profiling that uses word embeddings and a neural network to achieve high accuracy and speed.

ARG-miner is a crowdsourcing-based platform for curating antibiotic resistance genes, achieving high accuracy and efficiency compared to manual curation.

NanoARG identifies antimicrobial resistance elements from nanopore-derived environmental metagenomes, enabling analysis of ARGs, MRGs, MGEs, and pathogens.

This paper introduces deepARG, a deep learning method for predicting antibiotic resistance genes from metagenomic data, achieving high precision and recall over traditional methods.