Miniprotein dark-swan-lotus

Elise Therond / dark-swan-lotus

Type

Miniprotein

Binding

Myostatin (GDF-8)

Medium

Expressed

Yes

Ref

config_myostatin_0_model_4,

Collections

AGDF-8 Challenge Results

Sequence (86 AA)

EEEEERRREELERIIAAAEAAAEKALGRPLTPKEREEIRFTVEQIIQLASSKELRENGLDTSGKKISNLEVVDIAIKVSLQLIKLE

Computational Metrics

Experimental Metrics

Design Class

Miniprotein

Isoelectric Point

4.94pH

Molecular Weight

9.8kDa

Download CIF

Loading structure viewer...

Binding

Yes

Binding Strength

Medium, Strong

K_D

5.8×10⁻⁸M

k_on

1.4×10⁵M^-1s^-1

k_off

3.6×10⁻³s^-1

Download raw data

Kinetic Curves (SPR)

Concentration

10.0 nM

31.6 nM

100 nM

316 nM

1000 nM

Loading kinetic curves...

Design Method

This protein was designed using ETRFDiffusion3 + LigandPNN + ESM2 + AlphaFold + Selection

Backbone generation: we created a JSON file targeting several hotspots ( many of them for binding a very specific region). These are the targeted hotspots: "myostatin": { "dialect": 2, "infer_ori_strategy": "hotspots", "input": "./GDF8.pdb", "contig": "A41-375,/0,80-120", "select_hotspots": { "A170": "CA,CB", "A171": "CA,CB", "A172": "CA,CB", "A174": "CA,CB", "A186": "CA,CB", "A187": "CA,CB", "A188": "CA", "A189": "CA,CB", "A190": "CA,CB", "A191": "CA,CB", "A192": "CA,CB", "A193": "CA,CB", "A207": "CA,CB", "A228": "CA,CB", "A230": "CA,CB", "A232": "CA,CB", "A233": "CA,CB", "A234": "CA" }, "is_non_loopy": true } } Then , for ligandPmm, I tried to run one with a temperature of 0.1 and the other with 0.3 . We did no more than 10 sequence generations for each backbone.