You are here

Threading with Ligand

2 posts / 0 new
Last post
Threading with Ligand
#1

Hi everyone,

I'm trying to use the partial_thread application to fill in some gaps in a PDB crystal structure, 2hko.  This structure has an FAD ligand and a key water I'm trying to keep during threading.  I've cleaned the file as recommended, changed the FAD ligand atom types to match the canonical Rosetta atom types, and made similar changes to my water.  While the threading runs, the ligand gets thrown out.

This is the command I used:

partial_thread.static.linuxgccrelease -database $ROSETTA_HOME/main/database -in:file:fasta lsd1.fasta -in:file:alignment alignment.aln -in:file:template_pdb 2hko_cleaned.pdb -verbose

I'm pretty new to Rosetta so please bear with me!  Thanks,

-Eugene

AttachmentSize
2hko_cleaned.pdb405.63 KB
alignment.aln_.txt1.34 KB
lsd1.fasta_.txt677 bytes
Post Situation: 
Thu, 2018-12-06 15:10
ekwan

The output of the partial_thread application will only contain the contents of the input fasta file. If you don't have a ligand in the fasta file, it won't be present in the output.

I think you can take advantage of the fact that the partial_thread application actually takes "annotated sequence" fasta files, rather than strictly one letter ones. As such, you can add an entry like `Z[FAD]` to the sequence where you want your ligand to be. The Z comes from the one letter code specified in the IO_STRING line of the params file for your ligand (normally Z), and the entry in square brackets is the full name of the ligand (the name from the NAME line of the params file). You should then also include the ligand as a position in your alignment file, so partial_thread knows it's supposed to copy it over.

Sometimes this is too much hassle to get right, though. The potentially easier way of doing things is to do the partial threading on just the protein sequence, and then copy over the ligand coordinates after the fact. (You'll still need to do adjustments to the input fasta file to the hybridize stage, though, to make sure you get the ligand there.)

Fri, 2019-03-29 13:21
rmoretti