protein design

This paper develops novel base editors by diversifying TadA orthologs and using generative models to optimize specificity and efficiency.

This paper introduces AminoX, a platform for programmable covalent protein design using non-standard amino acids, enabling efficient screening and functional validation of covalent warheads and multifunctional proteins.

This paper designs synthetic transmembrane subunits to stabilize and expand membrane proteins, demonstrating their utility in structural biology and disease treatment.

The paper designs a binary fluorescent turn-on probe using a HaloTag mutant and a difluorinated xanthene ligand for covalent labeling and imaging applications.

This paper shows polyP regulates gene expression by modifying super-enhancer complex components, disrupting phase separation and impairing transcription factor function.

This paper introduces DeepUMQA-Global, a deep learning framework for estimating protein structure model accuracy, which outperforms existing methods in correlation metrics and consensus rankings.

RedNet improves protein binder design by incorporating side-chain information and contrastive decoding to enhance binding affinity, folding stability, and selectivity.

The paper identifies ABHD13 and ABHD16A as deacylases regulating SNAP25 family proteins and maps substrate selectivity to a single amino acid position.

The paper defines an extended translation initiation sequence (eTIS) and shows that a deep-learning model trained on eTIS features accurately predicts translation initiation, revealing that the Kozak sequence is not optimal and identifying conserved nucleotides that enhance initiation efficiency.

LysinFusion improves phage lysin prediction using a hybrid CNN-Transformer model, achieving higher accuracy and fewer false positives than existing methods.

ProteinFlux predicts protein conformational dynamics influenced by post-translational modifications using a generative framework, outperforming traditional methods in speed and accuracy.

This study uses synthetic nanobodies to identify functional regions of the Bacillus subtilis SMC complex, revealing the mid-region of the coiled coil as critical for DNA folding.

This study uses generative protein design to explore how sequences encode amyloid fibril structures, showing a continuous compatibility between sequences and a common amyloid fold.

This paper introduces a deep learning framework for designing proteins with site-specific post-translational modifications using latent diffusion models and a PTM-aware language model.

The paper develops a two-step in vitro evolution method to improve translational proteins and identifies mutations that enhance translation activity.

This paper designs a larger AAV capsid with T=3 symmetry using AI, increasing its cargo capacity to up to 35 kb.

This study compares deep learning models for generating antimicrobial peptides, finding that auto-encoders, especially Wasserstein auto-encoders, produce more diverse sequences than other models.

The paper audits structural biases in ML models for antimicrobial peptide design and finds that sequential training data lead to biased predictions due to uneven fold representation.

This study compares protein language models (pLMs) with simple baselines in combinatorial protein design, finding no consistent advantage for pLMs in predicting or generating superior variants.

The paper develops BERT-T6, a model using transfer learning on ProtBert embeddings to predict T6SEs with high accuracy and performance metrics.

This paper designs synthetic fusogens using machine learning, showing they enable efficient cell-cell and intracellular membrane fusion with higher activity than native SNARE complexes.

EctoShed is a gene delivery platform that enables expression of both membrane-bound and soluble proteoforms in adipocytes, revealing their functional roles in metabolic regulation.

This study evaluates refolding metrics in protein design, showing they are limited by evolutionary information and structural flexibility, reducing their reliability for filtering design candidates.

The paper uses GROQ-seq to profile function across a diverse protease library, showing robust activity across sequence divergence.

The paper shows how repositioning a gene within an operon changes the morphology and function of bacterial microcompartments.

The paper fine-tunes a protein language model using synthetic metalloprotein sequences to improve recovery of metal-binding motifs and identifies novel metal-binding sites.

This paper demonstrates that magnetic control of astrocytic calcium signaling can modulate neural circuit activity, introducing a tool for wireless, cell-specific calcium regulation.

This paper introduces Genie 3, a diffusion model that improves protein design by combining SE(3)-equivariance with hallucination, achieving state-of-the-art results in binder design and motif scaffolding.

DOMINO generates multidomain proteins by learning domain co-occurrence patterns and using them to design novel sequences with high structural confidence.

This study shows that encapsulating the enzyme TPI in HO BMC shells improves its thermal stability up to 62°C.