A5072055494- Enzyme Catalysis and Immobilization
- Protein Structure and Dynamics
- Microbial Metabolic Engineering and Bioproduction
- Bacterial biofilms and quorum sensing
- Computational Drug Discovery Methods
- Bacterial Genetics and Biotechnology
- Antimicrobial Peptides and Activities
- Machine Learning in Bioinformatics
- Peptidase Inhibition and Analysis
- Lipid Membrane Structure and Behavior
- Chemical Synthesis and Analysis
- Machine Learning in Materials Science
- Pancreatic function and diabetes
- Antibiotic Resistance in Bacteria
- Blind Source Separation Techniques
- Polyamine Metabolism and Applications
- Epigenetics and DNA Methylation
- Free Radicals and Antioxidants
- Monoclonal and Polyclonal Antibodies Research
- Biofuel production and bioconversion
- Enzyme Production and Characterization
- Amino Acid Enzymes and Metabolism
- Advanced Proteomics Techniques and Applications
- Topic Modeling
- Bioinformatics and Genomic Networks
Vanderbilt University
2021-2025
The Burkholderia cepacia complex (Bcc) is a group of bacteria including opportunistic human pathogens. Immunocompromised individuals and cystic fibrosis patients are especially vulnerable to serious infections by these bacteria, motivating the search for compounds with antimicrobial activity against Bcc. Ubonodin lasso peptide promising Bcc species, working inhibiting RNA polymerase in susceptible bacteria. We constructed library over 90 000 ubonodin variants 2 amino acid substitutions used...
Engineering enzymes to catalyze non-native substrates is critical for chemical synthesis. A significant challenge create a tool specialized in shifting an enzyme’s activity toward specified substrate. We developed SubTuner, physics-based computational that automates enzyme engineering catalyzing desired substrates. To test the performance of we designed three tasks – all aiming identify beneficial anion methyltransferase mutants synthesis S-adenosyl-L-methionine analogs: first conversion...
<title>Abstract</title> Lasso peptides (LaPs), characterized by their entangled slipknot-like structures, are a large class of ribosomally synthesized and post-translationally modified (RiPPs), with examples functioning as antibiotics, enzyme inhibitors, molecular switches. Despite thousands LaP sequences predicted bioinformatics, only around 50 distinct LaPs have been structurally in the past 30 years. Existing computational tools, such AlphaFold2, AlphaFold3 ESMfold, fail to accurately...
Hydrolase-catalyzed kinetic resolution is a well-established biocatalytic process. However, the computational tools that predict favorable enzyme scaffolds for separating racemic substrate mixture are underdeveloped. To address this challenge, we trained deep learning framework, EnzyKR, to automate selection of hydrolases stereoselective biocatalysis. EnzyKR adopts classifier-regressor architecture first identifies reactive binding conformer substrate-hydrolase complex, and then predicts its...
Hydrolases are a critical component for modern chemical, pharmaceutical, and environmental sciences. Identifying mutations that enhance catalytic efficiency presents roadblock to design discover new hydrolases broad academic industrial uses. Here, we report the statistical profiling rate-perturbing mutant with single amino acid substitution. We constructed an integrated structure−kinetics database hydrolases, IntEnzyDB, which contains 3907 kcats, 4175 KMs, 2715 Protein Data Bank IDs....
Data-driven modeling has emerged as a new paradigm for biocatalyst design and discovery. Biocatalytic databases that integrate enzyme structure function data are in urgent need. Here we describe IntEnzyDB an integrated structure–kinetics database facile statistical machine learning. employs relational architecture with flattened structure, which allows rapid operation. This also makes it easy to incorporate more types of data. contains kinetics from six commission classes. Using 1050 pairs,...
Cold-adapted bidomain enzymes are vital for transforming modern industries by decreasing energy consumption, delivering economic benefits, and fostering sustainability through reduced greenhouse gas emissions. Yet, the design strategies guiding their acquisition of cold adaptation remain unknown. Here, we developed an integrated computational-experimental strategy to engineer enhanced cold-adaptation. Using five model amylase variants exhibiting different degrees adaptation, identified a...
Lasso peptides (LPs), characterized by their entangled slipknot-like structures, are a large class of ribosomally synthesized and post-translationally modified (RiPPs), with examples functioning as antibiotics, enzyme inhibitors, molecular switches. Despite thousands LP sequences predicted bioinformatics, only around 50 distinct LPs have been structurally in the past 30 years. Existing computational tools, such AlphaFold2, AlphaFold3 ESMfold, fail to accurately predict structures due...
Lasso peptides (LaPs), characterized by their entangled slipknot-like structures, are a large class of ribosomally synthesized and post-translationally modified (RiPPs), with examples functioning as antibiotics, enzyme inhibitors, molecular switches. Despite thousands LaP sequences predicted bioinformatics, only around 50 distinct LaPs have been structurally in the past 30 years. Existing computational tools, such AlphaFold2, AlphaFold3 ESMfold, fail to accurately predict structures due...
The role of entropy in mediating the dynamic outcomes chemical reactions remains largely unknown. To evaluate change along post-transition state paths, we have previously developed entropic path sampling that computes configurational from an ensemble reaction trajectories. However, one major caveat this approach lies its high computational demand: about 2000 trajectories are needed to converge computation profile. Here, by leveraging a deep generative model, accelerated evaluates profiles...
Hydrolase-catalyzed kinetic resolution is a well-established biocatalytic process. However, the computational tools that predict favorable enzyme scaffolds for separating racemic substrate mixture are underdeveloped. To address this challenge, we trained deep learning framework, EnzyKR, to automate selection of hydrolases stereoselective biocatalysis. EnzyKR adopts classifier-regressor architecture first identifies reactive binding conformer an enantiomer-hydrolase complex, and then predicts...
Enzyme engineering techniques optimize enzymes to synthesize value-added chemicals, degrade environmental pollutants, and improve therapeutics. The field is entering a new era characterized by the increasing integration of computational strategies. While bioinformatics artificial intelligence (AI) have been extensively applied accelerate screening function-enhancing mutants, physics-based modeling methods, such as molecular mechanics quantum mechanics, serve essential complements in...
Engineering enzymes to catalyze non-native substrates is critical for chemical synthesis. A significant challenge create a tool specialized in shifting an enzyme’s activity toward specified substrate. We developed SubTuner, physics-based computational that automates enzyme engineering catalyzing desired substrates. To test the performance of we designed three tasks – all aiming identify beneficial anion methyltransferase mutants synthesis S-adenosyl-L-methionine analogs: first conversion...
Protein engineering holds immense promise in shaping the future of biomedicine and biotechnology. This review focuses on our ongoing development Mutexa, a computational ecosystem designed to enable "intelligent protein engineering". In this vision, researchers can seamlessly acquire sequences variants with desired functions as biocatalysts, therapeutic peptides, diagnostic proteins by interacting machine, similar how we use Amazon Alexa these days. The technical foundation Mutexa has been...
Data-driven modeling has emerged as a new paradigm for biocatalyst design and discovery. Biocatalytic databases that integrate enzyme structure function data are in urgent need. Here, we described IntEnzyDB an integrated structure-kinetics database facile statistical machine learning. employs relational architecture with flattened structure, which allows rapid operation. This also makes it easy to incorporate more types of data. contains kinetics from six commission classes. Using 1019...
Reaction dynamics trajectory simulations have been conducted to predict the product ratio of reactions with post-transition state bifurcation. However, it remains unknown how entropy reactive species along reaction path mediates ambimodal selectivity. Here, by leveraging deep generative model, we developed an accelerated entropic sampling approach that evaluates change post-transition-state for each using merely a few hundred dynamic trajectories. The new method, called bidirectional...
Abstract The Burkholderia cepacia complex (Bcc) is a group of bacteria including several opportunistic human pathogens. Immunocompromised individuals and cystic fibrosis patients are especially vulnerable to serious infections by these bacteria, motivating the search for compounds with antimicrobial activity against Bcc. natural product ubonodin lasso peptide promising Bcc species, working inhibiting RNA polymerase in susceptible bacteria. In this study, we developed high-throughput screen...
Hydrolase-catalyzed kinetic resolution is a well-established biocatalytic process. However, the computational tools that predict favorable enzyme scaffolds for separating racemic mixture are underdeveloped. To address this challenge, we trained deep learning framework, EnzyKR, to automate selection of hydrolases stereoselective biocatalysis. EnzyKR adopts classifier-regressor architecture first identifies reactive binding conformer an enantiomer-hydrolase complex, and then predicts its...
Abstract During intestinal inflammation, host nutritional immunity starves microbes of essential micronutrients such as iron. Pathogens scavenge iron using siderophores, which is counteracted by the lipocalin-2, a protein that sequesters iron-laden including enterobactin. Although and pathogens compete for in presence gut commensal bacteria, roles commensals involving remain unexplored. Here, we report Bacteroides thetaiotaomicron acquires inflamed utilizing siderophores produced other...
We reported the statistical profiling for rate-enhancing mutant hydrolases with single amino acid substitution. constructed an integrated structure-kinetics database, IntEnzyDB, which contains 3,907 experimentally characterized hydrolase kinetics and 2,715 Protein Data Bank IDs. The data involve 9% mutations. Mutation to nonpolar residues a hydrocarbon chain shows stronger preference rate acceleration than polar or charged residues. To elucidate relationship mutations, we categorized each...