All Competitors

A deep generative decoder that infers microRNA expression directly from bulk or single-cell mRNA gene expression via a shared mRNA/miRNA latent space.

A self-supervised masked autoencoder for RNA-seq count data, pretrained on 1.4M public samples to learn transferable transcriptomic representations without per-dataset re-training.

A constrained deep flow-matching framework for distributional translation of omics signatures across biological domains, such as mouse-to-human transcriptomics, without paired samples.

SE(3)-invariant masked autoencoder pretrained on ~370K AlphaFold-DB structures for protein fold representation learning, enabling frozen-feature and zero-shot fold classification.

A mechanistic-interpretability framework that trains sparse autoencoders on protein language model embeddings to extract interpretable features for zero-shot variant effect prediction.

A global protein structure tokenizer whose successive tokens add increasing detail, enabling adaptive-length representations, better generation, and zero-shot protein design.

A latent flow-matching method that repurposes protein language model embeddings to generate high-fitness protein variants without predictor guidance during sampling.

Bidirectional generative framework linking single-cell nuclear morphology and gene expression, built on a morphology foundation model trained on 21M+ segmented nuclei.

Pan-cancer multi-omic BERT-like foundation model that jointly encodes CpG-island DNA methylation and RNA-seq for zero-shot cancer classification and mutation prediction.

A fine-tuned scLDM variant trained on 14.5 million CD4+ T cells for counterfactual prediction of single-gene perturbation effects in immune cells.

A scalable latent diffusion model for generating realistic single-cell gene expression profiles, using a permutation-invariant VAE and flow-matching diffusion transformer.

Metric learning foundation model that embeds single-cell RNA-seq profiles into a unified space for scalable cell type annotation and cross-atlas similarity search across tens of millions of cells.

Contrastive masked ECG-text auto-encoder pretrained on paired electrocardiograms and clinical reports, enabling label-efficient and zero-shot cardiac diagnosis.

A variational autoencoder pretrained on 74 million human single-cell transcriptomes from the CELLxGENE Census for scalable batch correction, cell typing, and data integration.

A region-aware masked-autoencoder framework that learns self-supervised representations directly from fMRI time-series via per-ROI embeddings and graph attention.

Open respiratory acoustic foundation models pretrained on ~136K curated cough and breathing recordings for health tasks such as disease detection and lung function estimation.

Self-supervised vision transformer autoencoder pretrained on ~57,000 multi-contrast brain MRIs via masked image modeling for downstream brain tumor diagnosis.

Generative image perturbation autoencoder that predicts cellular morphological responses to chemical and genetic perturbations using untreated cell images as input.

Chromosome-wise explainable autoencoder for dimensionality reduction of DNA methylation data, achieving up to 400-fold compression while enabling interpretable CpG grouping analysis.

Explainable autoencoder for transcriptome analysis that identifies critical genes using SHAP-based attribution of neural network latent representations.

Self-supervised medical vision-and-language pretraining via multi-modal masked autoencoders that reconstruct masked image patches and text tokens.

Compositional Perturbation Autoencoder that predicts single-cell transcriptional responses to unseen drug combinations and doses using disentangled latent representations.

Variational autoencoder for single-cell RNA-seq that models raw count distributions directly, producing latent cell representations without normalization preprocessing.