protein design

This paper designs a water-soluble CD20 mimic using computational methods to enable binding studies and therapeutic development.

This paper introduces PEPIC, an explainable deep learning model for classifying plastic-hydrolytic enzymes from protein sequences, showing improved performance and biological validation.

The paper designs a panel of selective peroxygenases for terpene oxyfunctionalization using computational methods, resulting in improved enzymatic functions and new products.

The paper maps the metalloproteome of Deinococcus indicus DR1 by integrating structural and functional annotations to identify proteins involved in heavy-metal tolerance.

ProVADA adapts proteins to non-native contexts by steering a pretrained generator while preserving function, successfully converting renin and deimmunizing antibodies.

The paper compares LigandMPNN and directed evolution for redesigning the effector-binding site of the RamR transcription factor, finding limited overlap between the two methods' results.

Metalorian generates heavy metal-binding peptides using a diffusion model guided by a classifier, producing peptides with controlled specificity and reduced length compared to known proteins.

The paper introduces a multiplex method to screen CSP extracellular interactions using nanoparticles and fluorescent beads, showing improved sensitivity and dynamic range over ELISA-based assays.

This study identifies fluorescent tags for plant mannan synthases in yeast to accelerate polysaccharide biosynthesis research and reduce plant transformation time.

The paper identifies CRISPR-Cas12a variants with expanded PAM recognition, enabling targeting of ~25% of the human genome.

This study identifies the structural basis of RECQL5-induced RNA polymerase II transcription braking and reveals its coordination with the TCR complex for reactivation.

The paper develops a proximity labeling protocol to study FTL1 interactome in rice panicles, identifying proteins involved in trafficking and transcription, and showing FTL1's cell-to-cell mobility.

This paper uses a protein language model and structural representation tree to discover improved PET hydrolases, identifying KbPETase with higher thermal stability and degradation activity.

This paper introduces Chai-2, a model that designs antibodies with a 16% hit rate, achieving success in 50% of targets without prior binders, and enables rapid validation in under two weeks.

Treating low-confidence AlphaFold structures as noisy data at pLDDT-dependent diffusion times lets a 16.7M-parameter model generate long proteins with higher diversity, designability and novelty than Proteína.

The paper investigates the evolution and functional properties of Epsilon receptors in placozoa, revealing molecular determinants for ligand specificity and polyamine regulation.

This paper introduces FoldCraft, a fold-conditioned AlphaFold2-Multimer hallucination framework for de novo binder design, achieving higher structural and binding confidence than existing methods.

ProDualNet designs dual-target protein sequences by integrating sequence-structure information from two receptors, showing better recovery and success rates than existing methods.

SARITA is a large language model that generates synthetic SARS-CoV-2 Spike S1 sequences, accurately predicting mutations in future variants like Delta and Omicron.

This study uses directed evolution to extend the lifespan of an Arabidopsis enzyme by increasing its abundance and improving its resistance to a toxic intermediate.

STELLA is a self-evolving AI agent that improves through interaction, achieving high performance in biomedical tasks and driving experimental discoveries in oncology and protein engineering.

The paper develops a computational method to design proteins that bind small molecules tightly and specifically, validated experimentally with X-ray crystallography and functional assays.

This paper introduces an uncertainty-aware discrete diffusion model for protein design, improving sequence generation from 3D structures by incorporating position-specific uncertainty and pretrained models.

PRO-LDM generates protein sequences with tailored properties using a conditional latent diffusion model, enhancing diversity and design fidelity.

The paper introduces a co-design paradigm to enhance binding affinity between Ndh2 and quinones, demonstrating its application in improving biosensing capabilities.

This paper develops a three-in-one mucosal vaccine, 3Rs-NC, that elicits broad neutralizing antibodies against three ACE2-using sarbecovirus clusters and provides durable protection in mice.

The paper develops a high-throughput assay using a de novo fluorescent reporter to measure secreted protein via the Salmonella T3SS, revealing regulatory and structural insights.

This study introduces RelaVDEP, a virtual directed evolution pipeline using reinforcement learning to accelerate protein mutation discovery, showing improved catalytic activity in PETase through novel mutations.

This paper designs de novo Rhodamine Binder (Rhobin) tags with high brightness, stability, and applicability in live-cell imaging and extreme conditions.

This paper develops ELP bolaamphiphiles for constructing multifunctional globular protein vesicles, showing their structural and mechanical properties.