why cyclic peptide doesnt work for chain b

[MariaDeligian]

Hi , I am trying to follow the given example for macrocyclic binder but I was trying to make the binder cyclize in chain b, so what i provide is python $script
inference.output_prefix=$prefix
inference.num_designs=$num_designs
'contigmap.contigs=[12-14 A3-117/0 ]'
inference.input_pdb=$pdb
++inference.cyclic=True
diffuser.T=50
++inference.cyc_chains='b'
ppi.hotspot_res=['A51','A52','A50','A48','A62','A65']

I just change the order if contigs and the chain to be cyclic. the rest are exactly the same as the given example. however i cant get a cyclic binder with this. I am obviously specifying something wrong but havent figured out what yet. Could you please help me

[rclune]

I believe that even if you switch the order of the contigs (contigmap.contigs=[A3-117/0 12-14]) the binder will still be generated at the beginning of the target. This situation appears design_ppi.sh as well: the contigmap is contigmap.contigs=[A1-150/0 70-100] however the output has the binder as chain A.

The inference.cyc_chains option specifies which chain you want to make into a cyclic peptide if you are generating multiple peptides in your inference run.