REVIEW 18 cited by
Learning from Protein Structure with Geometric Vector Perceptrons
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Learning on 3D structures of large biomolecules is emerging as a distinct area in machine learning, but there has yet to emerge a unifying network architecture that simultaneously leverages the graph-structured and geometric aspects of the problem domain. To address this gap, we introduce geometric vector perceptrons, which extend standard dense layers to operate on collections of Euclidean vectors. Graph neural networks equipped with such layers are able to perform both geometric and relational reasoning on efficient and natural representations of macromolecular structure. We demonstrate our approach on two important problems in learning from protein structure: model quality assessment and computational protein design. Our approach improves over existing classes of architectures, including state-of-the-art graph-based and voxel-based methods. We release our code at https://github.com/drorlab/gvp.
Forward citations
Cited by 18 Pith papers
-
Contrastive learning of dynamical representations for enhanced molecular sampling
SelfTICA reformulates collective-variable discovery as contrastive dynamical representation learning on time-lagged data, decoupling feature learning from slow-mode extraction to produce reusable collective variables ...
-
EpiFormer: Learning Antigen-Antibody Interactions for Epitope Prediction via Geometric Deep Learning
EpiFormer improves epitope prediction F1 score by over 40% via early-fusion cross-attention in GNN layers and sparsity-aware objectives, while recovering known biology as emergent behavior.
-
ConTact: Contact-First Antibody CDR Design via Explicit Interface Reasoning
ConTact introduces a contact-then-act architecture with distance-biased cross-attention and contact-weighted loss for antibody CDR design, reporting 5-6% better backbone RMSD and superior contact metrics on CHIMERA-Be...
-
ConTact: Contact-First Antibody CDR Design via Explicit Interface Reasoning
ConTact decomposes CDR design into surface fingerprint learning, contact prediction, and contact-gated sequence generation using distance-biased attention and weighted loss, reporting 7% RMSD and 10% F1 gains on CHIME...
-
h-MINT: Modeling Pocket-Ligand Binding with Hierarchical Molecular Interaction Network
h-MINT improves ligand-protein binding affinity prediction by 2-4% and virtual screening metrics by 1-3% via overlapping fragment tokenization and hierarchical modeling.
-
BioBlobs: Unsupervised Discovery of Functional Substructures for Protein Function Prediction
BioBlobs discovers functional substructures unsupervised by learning to compress proteins into task-adaptive blobs from which function is predicted, recovering known catalytic sites from protein-level labels alone.
-
Align Your Structures: Generating Trajectories with Structure Pretraining for Molecular Dynamics
Structure-pretrained diffusion plus an equivariant temporal interpolator generates chemically realistic MD trajectories on small molecules, tetrapeptides, and proteins by separating spatial and temporal learning.
-
Scalable Inference-Time Annealing with Surrogate Likelihood Estimators
SITA performs scalable inference-time annealing of flow-based models on molecular systems by substituting energy-based surrogate likelihoods for divergence-based importance weights.
-
SurfDesign: Effective Protein Design on Molecular Surfaces
SurfDesign introduces surface-conditioned protein design via manifold modeling and equivariant message passing on surfaces integrated with pretrained language models, outperforming prior methods on binder and enzyme d...
-
AgForce Enables Antigen-conditioned Generative Antibody Design
AgForce improves antigen-conditioned antibody design by using framework dropout, gated bottlenecks, hyperbolic cross attention, MDN sequence head with Potts-like coupling, annealed MCL, and antigen cycle consistency t...
-
EvoStruct: Bridging Evolutionary and Structural Priors for Antibody CDR Design via Protein Language Model Adaptation
EvoStruct integrates evolutionary priors from a protein language model with structural priors from an E(3)-equivariant GNN to raise amino acid recovery by 16% and diversity by 2.3x on CHIMERA-Bench while cutting perpl...
-
Cross-Chirality Generalization by Axial Vectors for Hetero-Chiral Protein-Peptide Interaction Design
Axial-feature injection into an E(3)-equivariant latent diffusion model enables zero-shot D-peptide binder design, with one CD38 binder (KD ≈ 10 µM) validated in vitro.
-
BioBlobs: Unsupervised Discovery of Functional Substructures for Protein Function Prediction
BioBlobs compresses proteins into a small set of cohesive substructures and predicts function from these blobs alone, recovering catalytic sites from protein-level labels across multiple encoders.
-
Tokenizing Loops of Antibodies
Igloo is a multimodal antibody loop tokenizer that, when plugged into protein language models, modestly improves loop retrieval, affinity prediction, and structure-consistent loop generation.
-
Protein Representation Learning with Secondary-Structure and Energy-Filtered Hydrogen-Bond Graphs
Secondary-structure-aware GNN using energy-filtered hydrogen-bond edges improves protein representation learning on standard benchmarks.
-
A Triple-Modal Contrastive Learning Framework with Sequence, Graph, and 3D Features for Drug-Target Interaction Prediction
TriMod-DTI uses contrastive learning across 1D sequences, 2D graphs, and 3D structures to outperform prior DTI methods on three benchmarks.
-
Evolutionary Profiles for Protein Fitness Prediction
EvoIF integrates within-family and cross-family evolutionary signals into a compact model to achieve competitive or state-of-the-art zero-shot fitness prediction on ProteinGym using only 0.15% of typical training data.
-
STELLA: A Multimodal LLM for Protein Functional Annotation via Unified Sequence-Structure Encoding
STELLA aligns ESM3 bimodal sequence-structure encodings with Llama-3.1-8B text modeling to claim state-of-the-art results on protein functional description prediction and enzyme-catalyzed reaction prediction.
Discussion (0). Sign in to comment.