Adam P. Arkin

The paper updates the ModelSEED Biochemistry Database with expanded compounds and reactions, integrating multi-source thermodynamic data and grading reaction reliability.

This study identifies host factors critical for ssRNA and ssDNA phage infection in E. coli, revealing conserved dependencies and lineage-specific requirements.

The study identifies surface glycan layers as primary or optional recognition substrates for phage infection in E. coli O157:H7, revealing their role in receptor access and infection outcomes.

This paper constructs a phage-bacteria interaction atlas for Klebsiella pneumoniae and develops a genome-guided machine learning model to predict phage-host interactions, outperforming traditional methods.

This study shows that filtering genomes based on coherent genotype-phenotype relationships improves cross-dataset prediction of microbial carbon utilisation, outperforming larger but less coherent datasets.

This paper identifies the WalRK two-component system's role in maintaining cell envelope integrity by compensating for reduced capsule and wall teichoic acid attachment in Streptococcus pneumoniae.

This study shows that probiotic communities can prevent Burkholderia thailandensis infection in mice by reducing niche overlap with the pathogen.

The paper uses controlled enrichments along ecological gradients to uncover eco-evolutionary forces shaping microbial biodiversity in subsurface sediment.

The KBase Research Agent automates genome analysis workflows, producing reproducible results and annotated genomes without human intervention.

The paper develops a genome-based framework to predict bacteriophage infectivity against diverse Pseudomonas aeruginosa strains using machine learning and experimental data.

The paper integrates thirteen fungal metabolic models into a unified framework, improves their consistency, and uses them to analyze bioenergetic differences across species.

The paper predicts phage receptor specificity from genome data using large-scale genetic screens and machine learning, achieving high accuracy for E. coli phages.

This study shows that carbon storage polymers like PHB provide a fitness advantage to microbes during environmental transitions by prolonging proliferation and aiding recovery from starvation.

The paper introduces a Bayesian Personalized Ranking model to predict microbial substrate preferences and develops an agent that integrates this model with a growth model and LLM for autonomous microbial analysis.

The paper constructs a barcoded insertional mutant library in Trichoderma atroviride to enable high-throughput functional genomics in filamentous fungi.

This paper develops a machine learning framework to predict strain-level phage-host interactions from genomes, achieving high accuracy without phylogenetic constraints.

The paper introduces Droplet RB-TnSeq to study phage-host interactions by isolating single mutant cells with defined phage titers, identifying host factors beyond receptor-level interactions.

The paper introduces Dual Tn-seq to assess genome-wide genetic interactions in Streptococcus pneumoniae by analyzing double mutants, revealing new biochemical pathway components.

The paper identifies and validates novel plasmid origins of replication from an environmental Pantoea strain, enabling expression of heterologous proteins and pathways in both model and non-model bacteria.

The paper examines microbial community diversity in a uranium-contaminated aquifer, finding reduced taxonomic and phylogenetic diversity but stable functional diversity.

The paper identifies lactuchelins, novel lipopeptide siderophores from Pseudomonas lactucae, which inhibit Xanthomonas campestris pv. campestris 8004 through iron chelation.

This study identifies pathway partitioning as a mechanism for complete denitrification in microbial communities and shows how imbalances in substrate utilization can disrupt cooperativity and lead to nitrous oxide emissions.

The paper updates GapMind to include novel enzymes for amino acid biosynthesis, reducing pathway gaps in prokaryotes from 1.4 to 0.7 per genome.

The paper introduces Coaux-Seq, a method using DNA barcoded fragment libraries and auxotroph complementation to characterize protein function in bacteria, identifying 53 previously unvalidated proteins.

This paper uses high-throughput genetics to identify nutrient utilization and accessory energy metabolism genes in the methanogen Methanococcus maripaludis.

This study investigates how genomic and environmental factors influence the distribution of Castellaniella species in a contaminated subsurface, identifying genetic mechanisms underlying their adaptation to stress.

The paper introduces a web-based tool for fast comparative genomics analysis of bacteria and archaea, enabling quick identification of protein homologs and gene neighborhoods across diverse taxa.

The paper describes interactive tools to improve functional annotation of bacterial genomes by combining multiple resources to infer protein functions and bacterial capabilities.

ModelSEED v2 improves genome-scale metabolic model reconstruction with enhanced energy metabolism and better pathway curation, leading to more accurate and comprehensive models.

This study identifies two genes in a two-component system whose deletion increases PHB productivity in Cupriavidus basilensis under non-stress conditions.