Problems in the design of multi-epitope protein scaffolds

[gugu-dog]

Hello! I am very interested in your research. Therefore, I attempted to design a 5-epitope scaffold protein using Genie2. However, the output consisted only of discontinuous amino acid residues, and PyMOL did not display any recognizable protein structure. Is it possible to design a 5-epitope scaffold protein with Genie2? Below are my input commands, the original PDB file format, and one of the results visualized in PyMOL.

python sample_scaffold.py --name g2 --epoch 30 --scale 0.4 --outdir results/g2

REMARK 999 NAME   chainAgenie2
REMARK 999 PDB    chainAgenie2
REMARK 999 INPUT   5   30
REMARK 999 INPUT  A 248 259 A
REMARK 999 INPUT   5   30
REMARK 999 INPUT  A 565 580 A
REMARK 999 INPUT   5   30
REMARK 999 INPUT  A 748 760 A
REMARK 999 INPUT   5   30
REMARK 999 INPUT  A10151027 A
REMARK 999 INPUT   5   30
REMARK 999 INPUT  A10341048 A
REMARK 999 INPUT   5   30
REMARK 999 MINIMUM TOTAL LENGTH      100
REMARK 999 MAXIMUM TOTAL LENGTH      400

[yeqinglin]

Thank you for your interests! I assume you are doing multi-motif scaffolding here (with flexible relative positions and orientations among epitopes. If this is the case, you would need to set the last character of each scaffold definition line to a different character (to indicate different motif groups). Something like this:

REMARK 999 NAME   1PRW_four
REMARK 999 PDB    1PRW
REMARK 999 INPUT      5  20
REMARK 999 INPUT  A  21  32 A
REMARK 999 INPUT     10  25
REMARK 999 INPUT  A  57  68 B
REMARK 999 INPUT     10  25
REMARK 999 INPUT  A  94 105 C
REMARK 999 INPUT     10  25
REMARK 999 INPUT  A 130 141 D
REMARK 999 INPUT      5  20
REMARK 999 MINIMUM TOTAL LENGTH      88
REMARK 999 MAXIMUM TOTAL LENGTH      163