All Competitors

End-to-end RNA 3D structure prediction from sequence alone, coupling the RNA-FM language model with an Invariant Point Attention structure module.

Chromatin accessibility prediction across 29 human immune cell types, using in-silico saturated mutagenesis to score constrained regulatory regions.

Cellular senescence prediction from protein sequence, pairing ESM-2 embeddings with a hybrid BiLSTM-CNN classifier at 86.43% test accuracy.

Protein language model that captures short- and long-range residue co-evolution through a dual pre-training objective, at 3B parameters.

Frugal conditional diffusion model that samples human SNP haplotypes in PCA space, producing artificial genomes for a chosen continental ancestry.

SMILES transformer pretrained to predict 113 RDKit molecular descriptors, giving embeddings that carry physicochemical properties into ADMET models.

Masked-autoencoder foundation model that pre-trains a 3D Residual Encoder U-Net on roughly 39,000 brain MRIs for volumetric image segmentation.

Long-context protein language model on a bidirectional Mamba backbone, outperforming ESM-2 by up to 30% at matched training token budgets.

Protein language model that explains single-site mutation effects in natural language and proposes new mutants from free-text instructions.

Demultiplexer for direct RNA nanopore sequencing that basecalls the DNA barcode inside the RT adapter, reaching 99% precision on up to 96 barcodes.

Multimodal RNA chat model pairing a frozen RNA-FM encoder with a Llama-3 8B backbone to answer natural-language questions about a submitted sequence.

Lasso peptide language model that adapts ESM-2 to threaded RiPP core sequences, supplying embeddings for cyclase substrate and activity prediction.

Dual-language transformer pretrained on paired protein and mRNA coding sequences, scoring protein and mRNA properties and generating optimized CDS.

Contrastive dual-encoder aligning T-cell receptor CDR3 and peptide epitope sequences in one latent space to rank which receptors bind which antigens.

DNA methylation foundation model reconstructing genome-wide profiles from sparse input. Outperforms GrimAge2 aging clocks on mortality prediction.

Single-cell perturbation prediction model that adds gene-level language embeddings from NCBI, UniProt, and Gene Ontology to scGPT representations.

Language model over whole-night sleep stage sequences that corrects automated sleep staging and supplies features for sleep disorder diagnosis.

Reprograms a frozen single-target diffusion model for dual-target drug design by composing SE(3)-equivariant messages across two aligned pockets.

Peptide representation model that DoRA-tunes ChemBERTa on 100,000 modified and bioactive peptide SMILES for therapeutic property prediction.

Protein sequence design model that identifies each residue from the voxelized atomic microenvironment around it, reaching 68.33% accuracy on TS500.

Breast cancer recurrence risk read from an H&E slide and six routine clinical variables, using a frozen pan-cancer pathology encoder.

Multi-modal, multi-task biological foundation model trained on 2 billion samples spanning proteins, small molecules, and single-cell gene expression.

Self-supervised vision transformer pretrained on chest X-rays to produce a domain-specific foundation model for classification and lung segmentation.