protein design

BoltzDesign1 inverts AlphaFold3 to design protein binders for diverse targets without model finetuning, achieving high in silico success rates and structural diversity.

CodonTransformer optimizes DNA sequences for protein expression across species by learning from natural codon usage patterns and generating sequences with appropriate regulatory features.

This paper introduces a method to improve cancer prognosis by analyzing cell-type specific epigenetic alterations, finding better prognostic models in liver and kidney cancer.

This paper introduces a method to achieve 100% protein fusion yield using Connectase by addressing its reversibility in 1:1 conjugation reactions.

The paper introduces a method to control protein activity through dimerization-based conditional reconstitution, demonstrating its application in inducing cell death, virus detection, and immune cell stimulation.

This paper introduces PUSDA, a framework combining ultra-high throughput sorting and NGS to generate large-scale sequence-function data, enabling ML-driven enzyme engineering with improved efficiency and productivity.

The paper evaluates sequence and structural similarity metrics for predicting shared paralog functions using synthetic lethality and protein-protein interactions.

The paper designs artificial multidomain transmembrane transporters using a combined generative landscape and molecular dynamics approach, achieving native-like functionality.

This paper introduces LigandMPNN, a deep learning method for protein sequence design that models non-protein atomic contexts, achieving better performance than existing methods in recovering interacting residues.

The paper introduces irCLIP-RNP and Re-CLIP methods to identify dynamic RNA-protein assemblies, revealing new patterns of RBP-RNA associations and EGF-induced remodeling.

This paper computationally designs multi-component protein filaments that assemble into ordered fibers, validated by cryoEM, and shows regulation via phosphorylation.

This paper introduces a method using autoregressive generative models to reconstruct ancestral protein sequences while accounting for epistasis, outperforming existing methods in simulations and experimental data.

The paper develops a suite of pre-assembled Golden Gate vectors for efficient protein expression and purification in E. coli, enabling customizable and high-throughput protein engineering.

The paper improves a GABA sensor, iGABASnFR2, with higher sensitivity and faster kinetics, enabling better in vivo imaging of GABAergic transmission.

The paper introduces a method to efficiently integrate gene libraries onto yeast cytosolic plasmids using integrases, enhancing the versatility of OrthoRep for protein engineering.

This study compares generative AI models for protein design, finding structural models produce more confident structures but less diversity, while language models generate more diverse proteins with lower structural confidence.

The paper uses directed evolution to create a chloride-sensitive fluorescent protein indicator called ChlorOFF by mutating OFPxm.

The paper experimentally defines and models phase portraits of genetically encoded oscillators in yeast, showing how noise influences frequency and amplitude, enabling noise-guided design of synthetic protein dynamics.

The paper introduces a high-throughput proteomic platform called SPOC that enables real-time label-free screening of protein interactions with kinetic analysis capabilities.

The paper introduces a computational method to predict protein evolution by reconstructing a protein's evolutionary history and suggesting future amino acid substitutions to improve desired properties.

This paper describes a protocol to produce and purify the La domain of human LARP6, showing that RNA binding increases its stability and reduces aggregation.

This paper develops ATOM biosensors using mStayGold and mBaoJin to detect PSD95, Gephyrin, and HOMER1 with high specificity and turn-on ratios over 100-fold.

The paper demonstrates a computational strategy to design new PET hydrolases by replicating functional motifs from a template enzyme using deep learning and molecular computations, resulting in enzymes with high activity and thermostability.

The paper introduces AlphaPanda, an algorithm combining transformer, 3DCNN, and diffusion models for antibody sequence and structure design, achieving high-quality results with reduced training data.

This paper develops T-SenSER, a synthetic biosensor for T cells that enhances anti-tumor responses by sensing TME factors like VEGF and CSF1, showing improved efficacy in lung cancer and multiple myeloma models.

This paper introduces ProPPr, a method for proteomic profiling of phospho-tau aggregates in tauopathies, revealing common and disease-specific proteins associated with these lesions.

The paper investigates the structural interaction between HAQ STING and COPA, identifying a potential drug-binding pocket that could lead to therapeutic strategies for COPA disease.

This paper introduces AMPCLGPT, a model that generates antimicrobial peptides and predicts their MIC values using contrastive learning and pre-training.

CHIEF improves protein design by combining five models, achieving higher sequence recovery and lower perplexity, and successfully designing functional malate dehydrogenase.

This paper introduces T6CNN, a model that improves T6SE prediction by integrating signal sequences, evolutionary data, and pre-trained features, achieving higher accuracy, sensitivity, and specificity than existing tools.