Chains reordered when duplicate sequences are present

[jgreener64]

Thanks for making AlphaLink2 available. If I have an A2B2 complex with the FASTA as:

>A
MTRNIGLKVKEPKRECEDKNCPFHGELAIRGKLFDGKVTGSKAKQTITLQKDAPVYFNKFKRYARGKSSIHAHVPGCIDVESGDNVLTAECRPISKSVSYVVVEVRS
>B
MVVNLKAKKRLASRVTGVGVHRIKFDADHLTDVADAITRENIRSLITANTIKIDSFTGTSRGRAQTKKAQRNKRGTKQGSKQGRKGARVGKKEIYVAKVRSLRRLLKIAKDRKELTNPEFWALYKKVGGNTVRNKAHLRQLMEEIVAKRKN
>C
MTRNIGLKVKEPKRECEDKNCPFHGELAIRGKLFDGKVTGSKAKQTITLQKDAPVYFNKFKRYARGKSSIHAHVPGCIDVESGDNVLTAECRPISKSVSYVVVEVRS
>D
MVVNLKAKKRLASRVTGVGVHRIKFDADHLTDVADAITRENIRSLITANTIKIDSFTGTSRGRAQTKKAQRNKRGTKQGSKQGRKGARVGKKEIYVAKVRSLRRLLKIAKDRKELTNPEFWALYKKVGGNTVRNKAHLRQLMEEIVAKRKN

and cross-links as:

44 A 148 B 0.2
44 C 148 D 0.2

then the AlphaLink2 output PDB files reorder the chains, i.e. the FASTA representing the PDB file is:

>A
MTRNIGLKVKEPKRECEDKNCPFHGELAIRGKLFDGKVTGSKAKQTITLQKDAPVYFNKFKRYARGKSSIHAHVPGCIDVESGDNVLTAECRPISKSVSYVVVEVRS
>B
MTRNIGLKVKEPKRECEDKNCPFHGELAIRGKLFDGKVTGSKAKQTITLQKDAPVYFNKFKRYARGKSSIHAHVPGCIDVESGDNVLTAECRPISKSVSYVVVEVRS
>C
MVVNLKAKKRLASRVTGVGVHRIKFDADHLTDVADAITRENIRSLITANTIKIDSFTGTSRGRAQTKKAQRNKRGTKQGSKQGRKGARVGKKEIYVAKVRSLRRLLKIAKDRKELTNPEFWALYKKVGGNTVRNKAHLRQLMEEIVAKRKN
>D
MVVNLKAKKRLASRVTGVGVHRIKFDADHLTDVADAITRENIRSLITANTIKIDSFTGTSRGRAQTKKAQRNKRGTKQGSKQGRKGARVGKKEIYVAKVRSLRRLLKIAKDRKELTNPEFWALYKKVGGNTVRNKAHLRQLMEEIVAKRKN

Is there a way round this?

[lhatsk]

Sorry for the late response, I'm finishing up a paper. This is indeed annoying, looks like Uni-Fold is reordering things. I will take a look once I am finished.