I used my custom version of BindCraft modified such that you can give a partial structure as input and added an ipSAE loss that was up-weighted compared to the other losses. As input I used a known nanobody scaffold (https://www.nature.com/articles/s41594-018-0028-6), and allowed only the CDR loops to be redesigned. The starting structure was based on the placement of an existing Hemagglutinin-neuraminidase antibody binder (PDB: 8TQI) where only the heavy chain is in contact with the viral protein, making it more nanobody-like. The Nipah target protein was heavily truncated for more efficient iterations.
id: shy-cobra-sand

Nipah Virus Glycoprotein G
0.76
82.27
--
14.2 kDa
129
id: misty-panther-quartz

Nipah Virus Glycoprotein G
0.73
83.23
--
14.0 kDa
129
id: soft-bear-iron

Nipah Virus Glycoprotein G
0.57
80.12
--
14.3 kDa
129
id: silver-falcon-orchid

Nipah Virus Glycoprotein G
0.53
81.28
--
14.1 kDa
129
id: deep-tiger-ash

Nipah Virus Glycoprotein G
0.51
85.34
--
14.1 kDa
129
id: bright-panda-ruby

Nipah Virus Glycoprotein G
0.35
79.90
--
14.4 kDa
129
id: jade-cobra-fern

Nipah Virus Glycoprotein G
0.27
81.43
--
14.1 kDa
129
id: small-moth-jade

Nipah Virus Glycoprotein G
0.16
80.28
--
14.4 kDa
129
id: jade-jaguar-bronze

Nipah Virus Glycoprotein G
0.00
79.84
--
14.4 kDa
129
id: small-ox-rose

Nipah Virus Glycoprotein G
0.00
80.55
--
14.2 kDa
129