CryoZeta is a de novo macromolecular structure modeling tool that integrates cryo-EM density information with a diffusion-model-based structure prediction pipeline.

It jointly leverages sequence information and cryo-EM map features to generate atomic models consistent with experimental density, and is effective for proteins, protein-nucleic acid complexes, and nucleic acid-only systems at resolutions up to ~10 Å. CryoZeta supports complexes with a total sequence length of up to ~2,700 residues or bases. If the total length exceeds this threshold, the system automatically switches to "cycle mode." Please note that each individual chain must not exceed the ~2,700 length limit.

If encounter problems, please contact Daisuke Kihara (dkihara@purdue.edu)

---

Please simply click "Upload" when you filled all input fields.

cryo_em_map :

Please do not input 0, you must provide a contour to remove the outside very noisy regions

contour_level :


resolution :


fasta(protein) :


fasta(DNA) :


fasta(RNA) :


Job Note (Optional):
This note will help you distinguish jobs later.