All Competitors

Cross-modal protein encoder that aligns ESM-2 sequence embeddings with ProteinMPNN structure embeddings in a shared space for cross-modal retrieval.

Neural Hamiltonian flow for protein sequence generation with inference-time control over composition and net charge via analytical bias potentials.

Self-supervised SE(3) geometric pretraining for protein backbone generators, improving designability, motif scaffolding, and conformational ensembles.

Contrastive model built on a genomic language model that predicts physical protein-protein interactions across a microbial proteome in linear time.

Graph-transformer foundation model pretrained on 3M protein pockets and 5M molecules as E(3)-equivariant graphs for protein-ligand representation.

Contrastive antibody language model predicting antibody-antigen binding specificity from sequence with a dual-encoder, cross-attentive architecture.

Learning-to-rank variant effect predictor that aligns overlapping deep mutational scanning assays into an assay-agnostic tolerance measure.

Enzyme function prediction model that uses contrastive learning to assign the first three EC digits to enzymes with functions unseen during training.

Polarizable machine-learning interatomic potential extending MACE with long-range electrostatics, trained on 100M OMol25 DFT calculations.

Conditional variational autoencoder for antimicrobial peptide design that disentangles sequence, function, and length for independent control.

Protein evolution model that learns indel dynamics and epistasis from unaligned sequences, simulating trajectories that yield functional proteins.

Retrieval-augmented framework for de novo peptide binder design that conditions generation on retrieved, structurally aligned binding evidence.

Flow-matching generative model for peptide sequence design that learns the protein semantic distribution, fine-tuned for antimicrobial peptides.

Diffusion model that generates continuous-time, all-atom biomolecular trajectories, reproducing conformational kinetics far more cheaply than MD.

Protein language model that encodes sequences as discrete words from a learned vocabulary for zero-shot function inference and protein design.

Latent diffusion model that designs D-peptide binders against native L-protein targets, generalizing across chirality via axial vector features.

Universal all-atom machine-learning force field for molecular dynamics, with ab initio-level accuracy on solvated biomolecules of ~1,500 atoms.

Multimodal generative model predicting viral antigenic change zero-shot from disentangled evolutionary, physicochemical, and structural signals.

Protein-ligand foundation model that maps coarse-grained structural representations directly to binding affinity, running ~26x faster than Boltz-2.

Unified drug design engine for protein-ligand structure prediction, binding affinity estimation, and compound generation from Isomorphic Labs.

Protein-ligand scoring function that conditions probabilistic geometric potentials on language model priors to rank docked poses and binding affinity.

Protein structure tokenizer that encodes a whole structure globally, with each successive token adding detail for adaptive-length representations.

Protein backbone generator running a diffusion transformer over SaProt structural tokens, with an IPA token cache to speed up de novo design.

Protein structural homology search from sequence alone, embedding proteins so that structural similarity becomes a fast nearest-neighbor lookup.