Loading BioAtlas...
Loading BioAtlas...
Vergleichen Sie biologische Modelle, Docking-Engines und Berechnungs-Workflows Seite an Seite. Filtern Sie nach Zugangsmodell, Rechenbedarf und Kategorie, um in unter einer Minute zu einer verwertbaren Entscheidungstabelle zu gelangen.
| Attribut | ＋ Werkzeug hinzufügen | |||||
|---|---|---|---|---|---|---|
| Category | Biological Foundation Models | Biological AI Models | Biological AI Models | Biological AI Models | Biological AI Models | — |
| Subcategory | Protein Language Models | Protein Design | Protein Design | Protein Design | Protein Design | — |
| Company | University of Bayreuth | Pacesa Lab | Boltz Team | Baker Lab | Google DeepMind | — |
| Access mode | Open Source | Research code or publication | Research code or publication | Research code or publication | Research Preview | — |
| Licence conditions | Apache 2.0 | Check code and model weights separately | Check code and model weights separately | Check code and model weights separately | Research Publication / DeepMind Research | — |
| Compute requirements | Standard GPU or multi-core CPU for inference | Check selected checkpoint and runtime; no GPU catalog recommendation | Check selected checkpoint and runtime; no GPU catalog recommendation | Check selected checkpoint and runtime; no GPU catalog recommendation | GPU cluster for generative diffusion sampling and candidate filtering | — |
| Biological application | De novo protein sequence generation and enzyme design | Nicht erfasst | Nicht erfasst | Nicht erfasst | De novo binder design, targeted cancer therapeutics, viral inhibitors, and biosensors | — |
| Research workflow | Prompt with motif or sample unconditionally -> evaluate folding with AlphaFold/ESMFold | Nicht erfasst | Nicht erfasst | Nicht erfasst | Specify target protein structure and epitope binding site -> generate de novo binder sequences and 3D backbones | — |
| Evidence level | Peer-reviewed research (Nature Communications 2022) | Peer-reviewed primary method | Primary research preprint | Primary research preprint | DeepMind Research Whitepaper & arXiv preprint (2024) | — |
| Commercial eligibility | Open source release | Unknown, requires audit | Unknown, requires audit | Unknown, requires audit | Research publication & early academic collaboration | — |
| Price | Free Open Source | Check selected model licence | Check selected model licence | Check selected model licence | Research Preview | — |
| Availability ⓘ | Available on Hugging Face and GitHub | Public research method/code | Public research method/code | Public research method/code | DeepMind research publication and ongoing academic collaboration | — |
| Features | Autoregressive unconditional and conditional sequence generation, Produces sequences with natural secondary structure propensities and globular folding, Trained on UniRef50 with zero structural supervision required, Available on Hugging Face Model Hub with straightforward pipeline inference | AF2-guided optimization pipeline proposes protein binder designs for target structures., Candidate binder sequences, modeled protein complexes | Target-conditioned protein and peptide binder generation model as reported in its preprint., Binder sequence and structure candidates | All-atom protein structure diffusion conditioned on specific ligand or nucleic acid atomic context., Generated protein all-atom structures | Generates high-affinity protein binders for diverse target proteins without initial starting fragments, Trained on large macromolecular structure repositories and AlphaFold predictions, Demonstrates up to 300-fold higher binding affinities than prior computational design methods, Experimental success rates ranging from 9% to 88% across diverse target proteins in wet-lab biophysical assays | — |
| Entry type | AI Model | Research AI model / method | Research AI model / method | Research AI model / method | AI Model | — |
| AI role | Generative Sequence Modeling | Published biomedical AI method | Published biomedical AI method | Published biomedical AI method | De Novo Binder Design | — |
| Input data | Nicht erfasst | Target protein 3D coordinates; AF2/MPNN workflow and target interface constraints | Target context; design constraints | Target atomic geometry; designed region and constraints | Nicht erfasst | — |
| Output data | Nicht erfasst | Candidate binder sequences; modeled protein complexes | Binder sequence and structure candidates | Generated protein all-atom structures | Nicht erfasst | — |
| Validation | Nicht erfasst | Requires independent dataset-specific evaluation | Requires independent dataset-specific evaluation | Requires independent dataset-specific evaluation | Nicht erfasst | — |
| Type | Autoregressive protein transformer | Nicht erfasst | Nicht erfasst | Nicht erfasst | Generative protein binder design model | — |
| Intended use | Nicht erfasst | Nicht erfasst | Nicht erfasst | Nicht erfasst | Nicht erfasst | — |
| Compatibility | Nicht erfasst | Nicht erfasst | Nicht erfasst | Nicht erfasst | Nicht erfasst | — |
| Manufacturer | University of Bayreuth | Nicht erfasst | Nicht erfasst | Nicht erfasst | Google DeepMind | — |
| Integration evidence | https://huggingface.co/nferruz/ProtGPT2 | Nicht erfasst | Nicht erfasst | Nicht erfasst | https://deepmind.google/discover/blog/alphaproteo-generates-novel-proteins-for-biology-and-health-research/ | — |
| Laboratory handoff | Synthesized sequences experimentally confirmed to adopt stable folds | Nicht erfasst | Nicht erfasst | Nicht erfasst | Designed binder sequences are ordered as synthetic DNA for yeast surface display and SPR/BLI affinity validation | — |