Loading…
Every biological foundation model, evaluated and ranked by the bio.rodeo team
Showing 2065–2088 of 2565 models
Single-cell Hi-C model annotating chromatin loops, TAD-like domains, and A/B compartments from sparse contact maps with no dense imputation.
Protein half-life prediction from sequence and AlphaFold-derived structure tokens, separating short- from long-lived proteins and nominating degrons.
Peptide collision cross section predictor for ion-mobility proteomics, built on frozen protein language model features plus ion charge and mass.
Sequence-conditioned diffusion model that samples backbone conformational ensembles for intrinsically disordered proteins without MSAs or simulation.
Codon optimization model designing host-specific coding DNA for a target protein, conditioned on any of 164 organisms across all domains of life.
Single-cell foundation model pretrained on 21 million mouse scRNA-seq profiles, with ortholog conversion extending its use to human transcriptomes.
Graph attention foundation model for spatial transcriptomics that assigns spatial domains zero-shot across gene panels, tissues, and technologies.
A 3D UNETR++ transformer that enhances experimental cryo-EM density maps, improving map-model agreement and automated de novo structure modeling.
Multimodal contrastive model aligning clinical EEG with free-text reports, enabling zero-shot EEG classification from natural-language prompts.
Antibody language model that generates full-length human B-cell receptor heavy and light chains by fine-tuning a general protein language model.
Cas13d guide RNA efficiency predictor that pairs RNA-FM sequence embeddings with predicted secondary structure to score sgRNA on-target activity.
Non-coding variant effect prediction for DNA methylation, scoring the direction and magnitude of a SNP's effect on CpG sites up to 100 kbp away.
Histopathology tile encoder trained by supervised multi-task learning over 16 annotated tasks, matching self-supervised encoders on 6% of patches.
Weakly supervised histopathology foundation model pretrained on 60,530 whole-slide images for cancer detection, prognosis, and molecular prediction.
Digital pathology foundation model trading tile geometry for scale: 380-pixel tiles cut into 10-pixel patches on a 1.1B-parameter ViT-g backbone.
Chest X-ray foundation model that pairs masked image modeling with image-report contrastive alignment for zero-shot diagnosis and phrase grounding.
Structure-based Gene Ontology annotation that proposes candidate functional regions in a residue graph before reading them out as GO terms.
Open-source reproduction of AlphaFold 3 that predicts structures of proteins, DNA, RNA, and small-molecule ligands, including their mixed complexes.
Protein-peptide docking that builds all-atom complexes by diffusion, covering 92 residue types including post-translational modifications.
EEG-to-language foundation model that pairs a Q-Conformer encoder with a frozen LLM to decode coherent sentences from non-invasive brain recordings.
End-to-end framework predicting protein structure and mutational fitness from a single sequence, with five-fold faster inference than ESMFold.
Multi-task EEG foundation model that treats brain signals as a foreign language, pairing a text-aligned neural tokenizer with a GPT-2 backbone.
Multimodal contrastive model aligning protein structure and sequence with ligand conformation and graph to retrieve binders without docking.