All Competitors

Multi-label therapeutic peptide classifier over frozen ProtT5 embeddings, scoring 15 bioactivities from antimicrobial to quorum-sensing in one pass.

Molecular glue degrader activity prediction from SMILES and protein sequence, with two-stage cross-attention that follows E3-then-substrate binding.

DNA-binding residue prediction across folded domains and disordered protein regions, with contrastive training that suppresses cross-predictions.

RNA foundation model pretrained on 223 eCLIP experiments to predict base-resolution RBP binding, with frozen embeddings that transfer downstream.

Cardiac foundation model with one shared Transformer encoder for ECG, PPG, and PCG, aligning modalities in latent cardiac time via a learned delay.

DARPin binder design constrained to the ankyrin-repeat consensus grammar, pairing a fine-tuned inverse-folding model with two structure oracles.

m6Am modification site predictor that fuses frozen RNA-FM embeddings, a one-hot BiLSTM, and a typed RNA structure graph by AUC-weighted voting.

Medical imaging vision-language model for chest X-ray, CT and MRI that generates reports, localizes lesions and compares studies over time.

Infrared spectroscopy foundation model pretrained on 60 million simulated spectra, then adapted to real FTIR measurements of molecules and mixtures.

Self-supervised single-lead ECG encoder pretrained on ten-minute ambulatory windows, giving patient-consistent embeddings for rhythm detection.

Enzyme function assignment by embedding retrieval, fusing three frozen protein language models into a 256-D EC-aware search space.

Single-cell perturbation response prediction by conditional latent diffusion, trained on the Tahoe-100M atlas of 100 million drug-treated cells.

Single-cell metabolome inference from scRNA-seq, learned from spatially paired Visium and MALDI-MSI sections by multiple-instance learning.

Vision-language foundation model for coronary angiography that aligns six-view cine studies with procedural reports for zero-shot lesion assessment.

EEG foundation model that decodes by matching neural activity to label text embeddings, with one instruction-tuned checkpoint covering seven tasks.

Computational pathology model predicting ten lymphoma subtypes from H&E whole-slide images and ordering the matching immunohistochemistry panel.

All-atom structure prediction for complexes of proteins, DNA, RNA, and small molecules, using Min-SNR diffusion weighting and the Muon optimizer.

miRNA-target interaction model fusing five gated evidence experts, whose frozen representation transfers to siRNA efficacy prediction.

Protein stability predictor scoring ΔΔG for substitutions, multi-point mutations and indels from a folding model's latent structure representations.

Virtual-cell model that compresses a transcriptome into eight discrete tokens in a reasoning LLM's vocabulary, predicting module-level drug response.

Molecular property prediction model pretrained jointly on SMILES strings and 2D graphs, fusing the two views through bidirectional cross-attention.

Antibody language model fine-tuned to predict hydrophobic interaction chromatography retention from paired heavy and light chain sequence alone.

Post-translational modification site prediction over frozen ProtT5 embeddings, sharing one model across six PTM types via anchor-based flow matching.