T-coffee expresso

[Miriam Drori]

Hi,

I’m trying to run Expresso on several peptides using their 3D coordinates from local PDB files (these are already aligned peptides). From what I understood, the way to do this locally without using the online BLAST is to use:

    t_coffee my_sequences.seq my_templates.tmpl -method TMalign_pair

My `my_templates.tmpl` looks like this:

    >1cka_pep  _P_ 1cka_pep.pdb
    >1ckb_pep  _P_ 1ckb_pep.pdb
    >1gbq_pep  _P_ 1gbq_pep.pdb
    >1gbr_pep  _P_ 1gbr_pep.pdb
    >1h3h_pep  _P_ 1h3h_pep.pdb
    ...

And my `my_sequences.seq` file is:

    >1cka_pep
    PPPALPPKK
    >1ckb_pep
    PPPVPPRR
    >1gbq_pep
    VPPPVPPRRR
    >1gbr_pep
    SPLLPKLPPKTYKRE
    >1h3h_pep
    APSIDRSTKPA
    ...

However, I get this message for all the pairs:

    4165710 -- INFORMATION: Method TMalign cannot be applied to [1cka  _pep vs 1ckb_pep], proba_pair will be used instead

Despite that, if I run `TMalign` manually on those same two PDB files, it works fine.

Any ideas what could be causing this fallback? Could it be a formatting issue or something in how T-Coffee tries to parse the structures internally?

Thanks in advance!