protein design

ProChoreo designs protein binders using conformational ensembles and generative deep learning, showing improved performance in binding and structure quality.

The paper investigates how SMG5 licenses SMG6 endonuclease activity in NMD by forming a composite active site with SMG6 PIN domain.

This study fine-tunes a protein language model with generated CAR sequences to predict CAR-T cell activity, showing improved performance with optimized training parameters.

The paper designs CsgA variants with altered β-strand lengths to explore structural plasticity in microbial β-solenoids, revealing how strand length affects nanofiber stability and material properties.

The paper introduces a dataset and a graph-transformer model to predict IDR-protein binding affinity, achieving high accuracy and generalizing to designed binders.

The paper develops a rule-based model to generate synthetic mitochondrial targeting sequences in yeast and demonstrates their functionality in directing proteins to mitochondria.

The paper uses AI to generate cortisol-binding peptides for non-invasive stress detection, identifying high-affinity candidates from a large sequence library.

CrossAffinity predicts protein-protein binding affinity using a sequence-based cross-attention model, outperforming both structure-based and sequence-based methods on multiple test sets.

This paper develops a cell-free directed evolution platform using droplet microfluidics to rapidly evolve RNA polymerases with new functions and applications.

This paper introduces GRIPs, a scalable platform for PTM editing using chimeric molecules, demonstrating functionality in multiple systems and confirming specificity through proteomics.

PEINT simulates realistic protein evolution by modeling sequence changes over time, generating functional proteins that retain enzymatic activity.

This paper develops DROP-CARs, drug-regulated CAR-T cells that control cell-cell interactions using small-molecule switches, showing in vitro and in vivo functionality.

ProtFlow generates high-quality protein sequences by learning the full semantic distribution of protein space, outperforming existing methods in generating antimicrobial peptides with broad activity.

The paper analyzes SARS-CoV-2-human protein-protein interactions to show Spike S may target cytoskeletal proteins via coiled-coil domains.

This paper designs and expresses non-natural membrane-less RNA organelles in E. coli, enabling algorithmic control over interactions and condensate structure.

This paper identifies data leakage in pre-trained pLLMs used for protein-protein interaction inference and shows their limitations in generalization.

This paper introduces a method for fast de novo protein design using a structure-inspired alphabet and Metropolis sampling to generate sequences without known homologues.

MemConverter reprograms protein localization between membrane and aqueous environments using MCP-guided sequence design and AlphaFold2 refinement, achieving stable integration or solubility.

This paper investigates the temporal dynamics of discrete protein diffusion models, revealing a decoupling between structural alignment and property optimization, and identifies a structural collapse regime due to Coulombic repulsion.

This paper introduces Subtimizer, a computational pipeline for designing potent and selective kinase peptide substrates using AI tools, showing improved activity and selectivity in tested kinases.

The paper describes a nine-amino-acid degron derived from MTCH2's 3'UTR that enables tunable protein degradation via the ubiquitin-proteasome system.

The paper introduces a toolkit using protein language models to define 'protein words' for functional prediction, outperforming motif-based methods across multiple tasks.

The paper introduces AtomPaint, a full-atom diffusion model for designing protein/non-protein binding interactions, showing it outperforms random guessing in sequence recovery.

This paper introduces MaSIF-PMP, a geometric deep learning model for predicting protein-membrane binding sites, showing improved performance and validating predictions with molecular dynamics simulations.

The paper develops peptide-based compartments that can be delivered into cells to sequester targets and degrade them, demonstrating potential for synthetic organelles in cell engineering.

This paper develops a human-based translational activator to restore protein expression in a Dravet syndrome mouse model, showing improved survivability and seizure threshold.

This paper introduces an AI-native biofoundry that uses natural language and active learning to autonomously engineer enzymes, achieving high hit rates and improved sequencing performance.

The paper describes methods to make a membrane-bound glycosyltransferase water-soluble for easier study and application in synthetic glycobiology.

The paper develops methods to improve the solubility and stability of intrabodies by optimizing linker design, domain orientation, and charge properties, enabling effective intracellular targeting of cytoplasmic proteins.

Biocentral provides an open embedding-based service for protein predictions, demonstrating its utility with a virus database analysis and training module.