protein design

Closed-Loop Multi-Objective Optimization for Receptor-Selective Cell-Penetrating Peptide Design

The paper develops a closed-loop framework for designing receptor-selective cell-penetrating peptides using multi-objective optimization and evaluates its effectiveness through experimental validation.

De novo design of allosterically controlled enzymes

This study designs allosterically controlled enzymes using denoising diffusion models, showing that activity can be activated or repressed by designed effectors.

Self-Architecting Protein Transformers: An Empirical Study

This paper evaluates a self-architecting transformer for protein representation, showing it achieves better performance than existing models with a smaller dataset.

EukaUTR: a foundation model unifying functional modelling and design of eukaryotic 3' UTRs

EukaUTR is a foundation model for eukaryotic 3' UTRs that predicts function and generates new sequences with regulatory properties.

Evolutionary analysis of PINK1 reveals key roles of the N- and C-terminal extensions and TOM20 in folding its kinase domain

This study identifies evolutionary differences in PINK1 orthologues that affect stability and activity, revealing the role of NTE, CTE, and TOM20 in kinase function and pathogenic variants.

PKProbDesign: RNA inverse folding including pseudoknots by optimizing thermodynamic folding probability

PKProbDesign optimizes thermodynamic folding probability for pseudoknotted RNA targets, achieving high pseudo-joint folding probability on most tested targets.

Specificity-driven protein binder design with Odin-Multi

Odin-Multi is a binder design framework that optimizes sequences against multiple targets simultaneously, improving cross-reactivity and specificity compared to single-target methods.

De novo Rubisco design with protein language models

This paper designs novel Rubisco variants using protein language models, identifying active candidates with improved CO2/O2 specificity and structural validity.

Expanding Macrocyclic Topology through Cysteine-to-N-Terminal Cyclisation Enables Covalent Peptide Inhibitor Discovery

This paper introduces a new cyclisation strategy using a bromomethyl picolinaldehyde linker to expand macrocyclic peptide topologies, leading to more potent and diverse inhibitors.

Mutation E300 recommended by protein language models gives ChrimsonR amplified photocurrent response

The paper uses protein language models to predict ChrimsonR mutations that enhance photocurrent response, validating two mutations that significantly improve function.

Protein Design Viz (PDV): lightweight protein structure visualization studio with validated quantitative analytics and antibody-specific toolkit

PDV provides a lightweight, browser-based molecular visualization tool with validated quantitative structural analysis and antibody-specific features.

Design and structure of protein cages based on helical fusion and machine learning

The paper designs and validates self-assembling protein cages using helical fusion and machine learning, achieving precise structures with high resolution and molecular mass accuracy.

Advancing Knotted Protein Design with ESM3: Guided Generation and Topological Insights

This study uses ESM3 to generate knotted proteins, finding high success rates with guided decoding and robustness in real knotted proteins.

De Novo Design and AlphaFold3 Evaluation of Protein Binders Targeting Specific Sites of MAP4K4

The paper designs small protein binders targeting specific MAP4K4 sites using computational tools and evaluates their binding using AlphaFold3, finding high-confidence candidates with novel sequence space solutions.

MECP2 MBD-ID Module: A Unified DNA/RNA Binding Interface Disrupted in Rett Syndrome

This paper identifies the MBD-ID module as a unified DNA/RNA binding interface in MECP2, showing it is disrupted in Rett syndrome due to mutations in the intervening domain.

The fitness landscape of a Form II Rubisco in a photosynthetic bacterium guides engineering of oxygen tolerance

The paper screens variants of a Form II Rubisco to improve oxygen tolerance, identifies key mutations, and uses a transformer model to guide engineering.

Development of approaches and identifying limitations towards functional yeast surface display of FLS2

The paper develops methods to improve yeast surface display of FLS2 for detecting bacterial ligands, identifies glycosylation as a limiting factor, and shows how modifying glycosylation sites enhances binding specificity.

Design and experimental characterization of specificity-switching mutational paths of WW domains

The paper designs mutational paths for WW domains to switch specificity using computational methods and validates them experimentally, showing high success rates for structured paths over random ones.

CatESO: Differentiable Enzyme Sequence Optimization Guided by Substrate-Aware kcat Prediction

CatESO optimizes enzyme sequences to improve catalytic turnover by integrating kcat prediction with differentiable design, achieving higher predicted activity and structural confidence than existing methods.

Towards Sparse Causal Features for Zero-shot Mutation Effect Prediction in a Protein Language Model

The paper introduces a sparse feature circuit framework to identify causal features in ESM-2 for zero-shot mutation effect prediction, showing that shared circuits explain mutation effects through 3D contact residues.

GRASP: A PLANT TRANSFORMATION-INDEPENDENT CRISPR-BASED SYSTEM FOR AFFINITY PURIFICATION OF SPECIFIC CHROMATIN LOCI

GRASP enables transformation-independent CRISPR-based chromatin isolation in plants by using dCas9-gRNA complexes to capture specific genomic regions from purified nuclei.

De novo design of ligand binding proteins using large language models alone

This paper demonstrates that large language models can generate de novo protein sequences that bind specific ligands without copying existing sequences, with a 25% hit rate in experimental validation.

Sequence-free landscape inference for directed evolution

This paper introduces SLIDE, a method to estimate landscape ruggedness in directed evolution using population-level phenotypic data and mutation rate, improving selection strategies for protein design.

Conjugative serine recombinase engineering enables genome integration in transformation-resistant bacteria and bioproduction pathway prototyping

This paper introduces cSAGE, a conjugation-based genome engineering platform that enables site-specific integration in transformation-resistant bacteria, demonstrating its utility in bioproduction pathway prototyping.

Multiobjective learning and design of bacteriophage specificity

This paper uses deep learning to design bacteriophage receptor binding proteins with multiobjective specificity, showing high tunability and success rates in targeting diverse strains.

Parkinson's disease-linked D620N mutation selectively alters the brain-specific protein interactome of VPS35

This study identifies a selective effect of the D620N VPS35 mutation on the brain-specific protein interactome in mammalian cells and brain tissue.

Structural insights and engineering of deep-sea halophilic PET hydrolytic enzymes

The paper structures and engineers deep-sea halophilic PET hydrolases to improve their salt tolerance and catalytic efficiency for industrial applications.

A Metabolic Labeling Strategy for Tracking Protein Synthesis in Complex Biological Systems

This paper introduces TeACAT, a method to track newly synthesized proteins in mice by incorporating a non-canonical amino acid, enabling visualization and quantification with minimal perturbation.

RegimeFormer: A Large Protein Model of Global Perturbation Regimes

RegimeFormer identifies protein perturbation regimes using a large-scale model and atlas, improving prediction accuracy and enabling downstream applications.

Engineering a highly active thermophilic F1-ATPase by homolog-guided exploration and machine-learning-assisted prioritization

This paper engineers a thermophilic F1-ATPase with higher catalytic activity using homolog-guided mutation identification and machine learning, finding a 1.8-fold increase in rotation rate.