All Competitors

Cell world model pretrained on a 2.4M-cell mouse embryonic atlas, predicting one-step transcriptional state transitions and perturbation response.

Generative model that reconstructs single-cell spatial coordinates from scRNA-seq guided by spatial transcriptomics, without cell-type labels.

700M-parameter DNA language model pretrained on the rice pangenome, serving as a reusable base model for crop genomics and molecular breeding.

Lung pathology foundation model adapted from Virchow2 on whole-slide images, validated across 32 tasks spanning the lung diagnostic workflow.

Contrastive promoter-protein pretraining that aligns bacterial promoters with their encoded proteins to learn regulatory genomics representations.

Variant effect predictor pairing a protein language model with family-specific evolutionary constraints to score stability, binding, and epistasis.

Mixture-of-Experts genomic foundation model for the human microbiome, with 4.7B parameters pretrained on bacterial, archaeal, and phage genomes.

RNA language model that predicts secondary structure of internal ribosome entry sites from sequence alone, trained on roughly 50,000 IRES sequences.

Spatial transcriptomics foundation model for the tumor microenvironment, giving TME-aware embeddings and in silico perturbation from one checkpoint.

Metabolomic foundation model pretrained on UK Biobank NMR metabolite profiles, reused with a frozen backbone for aging, subtyping, and disease risk.

SE(3)-invariant masked autoencoder that learns protein fold representations from AlphaFold-DB structures, supporting zero-shot fold classification.

Sparse autoencoders trained on protein language model embeddings to expose interpretable features and drive zero-shot variant effect prediction.

Protein conformational ensemble tokenizer that learns a discrete alphabet of states from molecular dynamics, reusable as a frozen feature layer.

Conditional discrete diffusion model for protein variant generation, with a calibrated identity dial controlling drift from a wild-type sequence.

Signed heterogeneous graph foundation model over the SIGMA-KG knowledge graph, predicting drug mode of action and drug-drug interactions zero-shot.

Transformer tractography model for mouse-brain diffusion MRI, guided by axonal priors learned from Allen Mouse Brain Connectivity Atlas streamlines.

Microbiome world model that treats a community as a set of taxa, scoring how well each member fits and predicting community dynamics zero-shot.

Discrete diffusion model for conditional antibody sequence design with germline-absorbing noising that focuses learning on somatic variation.

Protein sequence embedding model, contrastively fine-tuned from ESM-2, that places functionally and structurally related proteins close together.

DNA language model combining Mamba state-space layers, gated dilated convolutions, and Fourier attention to capture multi-scale regulatory patterns.

Microbiome foundation models that treat microbial community composition as a language, enabling zero- and few-shot transfer across prediction tasks.

Self-supervised brain MRI foundation model built on DINOv3, pretrained on roughly 6.6 million unlabeled axial slices for neuroimaging tasks.

Direction-aware foundation model trained on bulk RNA-seq differential-expression profiles to simulate coordinated gene dynamics in viral infection.

Transcriptome foundation model for precision oncology, generalizing zero-shot across tissue, plasma cfRNA, and tumor-educated platelet modalities.