kiharalab/DiffModeler

DiffModeler: a diffusion model based protein complex structure modeling tool.

31
/ 100
Emerging

DiffModeler helps structural biologists and biochemists automatically build complete protein complex structures from cryo-electron microscopy (cryo-EM) data. You input a cryo-EM map (an image representing electron density) and, optionally, individual protein chain structures or sequences. The tool then outputs a high-accuracy 3D model of the assembled protein complex, even when the cryo-EM data is at challenging intermediate to low resolutions.

Use this if you need to accurately model the 3D structure of multi-chain protein complexes from cryo-EM maps, especially at intermediate (5-10 Å) or even lower (10-20 Å) resolutions where traditional methods struggle.

Not ideal if you are working with extremely high-resolution cryo-EM maps (better than 2 Å) where diffusion model-based backbone tracing might not be necessary, or if you lack access to GPU hardware.

structural-biology cryo-EM protein-modeling biochemistry macromolecular-assembly
No License No Package No Dependents
Maintenance 6 / 25
Adoption 8 / 25
Maturity 8 / 25
Community 9 / 25

How are scores calculated?

Stars

58

Forks

5

Language

Python

License

Last pushed

Jan 12, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/kiharalab/DiffModeler"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.