protein
PulseAugur coverage of protein — every cluster mentioning protein across labs, papers, and developer communities, ranked by signal.
2 day(s) with sentiment data
-
AI sequence models find new applications in genomics
Sequence models, similar to those used in language processing, are being applied to genomics to analyze protein and nucleotide sequences. These models learn evolutionary patterns by predicting masked portions of sequenc…
-
Machine learning accurately classifies complex link topologies
Researchers have developed a machine learning approach to classify complex link topologies, relevant to fields like polymer melts, DNA, and proteins. A feedforward neural network trained on a "writhe density matrix" ach…
-
New multimodal model MKB unifies scientific domains for AI-driven discovery
Researchers have introduced Monkey King Bang (MKB), a novel multimodal foundation model designed for scientific discovery across diverse domains. MKB utilizes a shared Transformer backbone with specialized components fo…
-
Discrete diffusion models optimized for molecular tasks
Researchers have explored the design space of discrete diffusion models for molecular optimization, focusing on how to adapt a pretrained generative model using a limited oracle budget. Their studies across various mole…
-
New 'Proto' language streamlines AI-driven generative biology design
Proto is a new domain-specific programming language designed for generative biology, enabling scientists to define biological systems by specifying goals and rules rather than selecting static parts. This approach allow…
-
New attacks reveal significant privacy risks in Graph Neural Networks
Researchers have developed two new generative reconstruction attacks, the graph-label conditioned (GLC) attack and the embedding-label conditioned (ELC) attack, to probe the privacy vulnerabilities of Graph Neural Netwo…
-
New Python ecosystem standardizes biomolecular sequence models
A new open-source Python ecosystem called MultiMolecule has been developed to standardize and facilitate the reuse of biomolecular sequence models. It provides a modular framework for handling RNA, DNA, and protein sequ…
-
Post-training stages critically shape biological reasoning models, study finds
A new study has investigated how different post-training stages impact the performance and generalization capabilities of biological reasoning models. Researchers trained over 100 models across genomics, transcriptomics…
-
Post-training stages critically shape biological reasoning models' generalization
A new study on over 100 biological reasoning models reveals that post-training stages significantly impact generalization capabilities. Continued pre-training aligns models with biological language, while supervised fin…
-
AI model DeepRHP aids design of protein-mimicking heteropolymers
Researchers have developed DeepRHP, a hybrid variational autoencoder designed to aid in the creation of synthetic random heteropolymers that can mimic protein functions. This model uses a semi-supervised framework, inco…
-
New DeCAF framework speeds up biomolecular structure generation
Researchers have developed a new framework called DeCAF to accelerate the process of generating 3D biomolecular structures. This method distills existing all-atom cofolding models into more efficient flow maps, signific…
-
Nanobodies engineered to light up cellular processes
Researchers have developed a novel method using nanobodies, which are smaller antibody fragments, to visualize processes within living cells. These specially engineered nanobodies are tagged with fluorescent labels and …
-
Traditional ML and Deep Learning Tied in Protein Structure Classification
A new study on arXiv compares traditional machine learning (ML) with deep learning (DL) for protein structure classification using dynamic graph representations. The research found that for most datasets, traditional ML…
-
AI maps 1.1 billion proteins, accelerating drug discovery
Artificial intelligence has been used to map over 1.1 billion proteins, creating a comprehensive resource for biological research and drug discovery. This extensive dataset is expected to accelerate scientific understan…
-
New AI models generate functional RNA sequences for protein interaction
Researchers have developed Moirain, a new suite of models for generating RNA sequences that can interact with specific proteins. This approach uses multimodal supervised fine-tuning and Direct Preference Optimization, b…
-
EPFL AI generates complete protein models and their dynamics
Researchers at EPFL have developed a neural network capable of generating complete all-atom models of proteins. This AI-driven approach also captures the dynamic movements essential to protein function, significantly st…
-
Deep learning revolutionizes single-cell sequencing for biological discovery
Deep learning techniques are playing a crucial role in advancing single-cell sequencing (sc-seq) technologies, which allow for the detailed analysis of individual cells. Sc-seq methods, recognized as a significant advan…