All Competitors

Gene representation framework fusing DNA, transcript, protein, text, and single-cell embeddings into one latent space that survives missing views.

Transcriptomics foundation model from Recursion that masks and reconstructs RNA-seq gene expression counts to learn reusable sample embeddings.

Flow-matching framework that translates omics signatures across biological domains, such as mouse to human transcriptomics, without paired samples.

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.

Vision-language foundation model linking human brain activation maps and neuroscience text for text-to-brain and brain-to-text generation.

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

Latent flow-matching method that repurposes protein language model embeddings to generate high-fitness protein variants without predictor guidance.

Bidirectional generative framework linking nuclear morphology and gene expression, built on a morphology model trained on 21 million segmented nuclei.

Pan-cancer multi-omic foundation model encoding CpG-island DNA methylation and RNA-seq for zero-shot cancer classification and mutation prediction.

Multi-target drug discovery framework pairing a diffusion-transformer generator with evolutionary latent-space search and synthesis-aware scoring.

Latent diffusion model for generating single-cell gene expression profiles, pairing a permutation-invariant autoencoder with a diffusion transformer.

Single-cell latent diffusion model fine-tuned on 14.5 million CD4+ T cells to simulate transcriptomic effects of single-gene perturbations.

Multimodal LLM that tokenizes single cells into discrete VQ-VAE codebook tokens, letting one model reason jointly over transcriptomes and text.

Protein structure autoencoder compressing backbone coordinates into a latent space, paired with a latent diffusion model for generative design.

De novo ligand design framework that generates protein binders and small molecules by inverting gradients through a differentiable docking model.

All-atom protein representation model that learns from each residue's strictly local atomic neighborhood, capturing side-chain geometry and chemistry.

Flow-based protein structure tokenizer, a diffusion autoencoder replacing SE(3)-invariant components with global coordinates and standard attention.

scVI variational autoencoder trained on the Tahoe-100M drug-perturbation atlas, giving a 10-dimensional embedding of treated cancer cell states.

Single-cell foundation model trained by metric learning to embed scRNA-seq profiles for cell type annotation and similarity search in cell atlases.

ECG foundation model pretrained on 12-lead waveforms paired with clinical reports, enabling label-efficient and zero-shot cardiac diagnosis.

Variational autoencoder pretrained on 74 million human single-cell transcriptomes from the CELLxGENE Census for batch correction and cell typing.

Self-supervised masked autoencoder for functional MRI that learns representations from BOLD time-series with per-ROI embeddings and graph attention.

Respiratory acoustic foundation models pretrained on roughly 136K cough and breathing recordings for disease detection and lung function estimation.