- Posts: 43
Martinize.py - missing residues in the created itp files
- z.sichaib
-
Topic Author
- Offline
- Senior Boarder
Less
More
4 years 1 month ago #8454
by z.sichaib
Martinize.py - missing residues in the created itp files was created by z.sichaib
Dear Martini users,
I am having a bizarre result when using martinize.py (version 2.6).
First I run an unbiased CG trajectory on my system using martini 22 FF.
The next step is to use the last frame from the unbiased trajectory and run CG metadynamics using Elnedyn FF.
For that, I extracted the last frame from the unbiased CG trajectory and backmapped it using initram and backward (from martini to amber). Next, I used the resulting AA structure as input to martinize.
My structure is a dimer with 165 residues. From the output of Martinize, the resulting itp file for chain B misses 2 residues (stops at 163) but for chain A everything seems correct. The resulting CG pdb structure also misses atoms in both chains!
I tried using a structure from another frame, different format of the structure (gro and PDB) but still not working. I don't really see where is the problem.
I put all the files here, in case you want to reproduce the problem.
www.dropbox.com/sh/fb5imf64anytj60/AABoC...k0wCj59WXVgLNDa?dl=0
aa_amber.gro ---> aa structure from the backmapping and for which I would like to generate a topology with elnedyn FF.
The rest of the files are outputs of martinize:
martinize.py -f aa_amber.gro -o topol.top -x CG.pdb -dssp /home/dssp/dsspcmbi -p backbone -ff elnedyn22
I would really appreciate your help with this problem.
I am having a bizarre result when using martinize.py (version 2.6).
First I run an unbiased CG trajectory on my system using martini 22 FF.
The next step is to use the last frame from the unbiased trajectory and run CG metadynamics using Elnedyn FF.
For that, I extracted the last frame from the unbiased CG trajectory and backmapped it using initram and backward (from martini to amber). Next, I used the resulting AA structure as input to martinize.
My structure is a dimer with 165 residues. From the output of Martinize, the resulting itp file for chain B misses 2 residues (stops at 163) but for chain A everything seems correct. The resulting CG pdb structure also misses atoms in both chains!
I tried using a structure from another frame, different format of the structure (gro and PDB) but still not working. I don't really see where is the problem.
I put all the files here, in case you want to reproduce the problem.
www.dropbox.com/sh/fb5imf64anytj60/AABoC...k0wCj59WXVgLNDa?dl=0
aa_amber.gro ---> aa structure from the backmapping and for which I would like to generate a topology with elnedyn FF.
The rest of the files are outputs of martinize:
martinize.py -f aa_amber.gro -o topol.top -x CG.pdb -dssp /home/dssp/dsspcmbi -p backbone -ff elnedyn22
I would really appreciate your help with this problem.
Please Log in or Create an account to join the conversation.
- peterkroon
-
- Offline
- Gold Boarder
Less
More
- Posts: 210
4 years 1 month ago #8455
by peterkroon
Replied by peterkroon on topic Martinize.py - missing residues in the created itp files
The coarse-grained protein_b.itp also contains only 163 residues.
Please check the itps/pdbs of your first CG simulation carefully, and if that does not provide insights, please post the outputs and command lines from backwards and Martinize.
Please check the itps/pdbs of your first CG simulation carefully, and if that does not provide insights, please post the outputs and command lines from backwards and Martinize.
Please Log in or Create an account to join the conversation.
- z.sichaib
-
Topic Author
- Offline
- Senior Boarder
Less
More
- Posts: 43
4 years 1 month ago #8456
by z.sichaib
Replied by z.sichaib on topic Martinize.py - missing residues in the created itp files
Thank you for your reply.
In my first CG simulation, everything seems fine, I have 165 residues in both chains.
The resulting backmapped aa structure (aa_amber.gro) also seems fine with 165 residues on both chains. Even when I put it as an input to Martinize, the output of this command line is:
martinize.py -f aa_amber.gro -o topol.top -x CG.pdb -dssp /home/dssp/dsspcmbi -p backbone -ff elnedyn22
INFO Found 2 chains:
INFO 1: A (Protein), 2581 atoms in 165 residues.
INFO 2: B (Protein), 2581 atoms in 165 residues.
but then the resulting CG structure is missing residues. And the Itp files of (only) chain B misses 2 residues.
The command line of backward is:
./initram.sh -f last_frame.gro -o aa_amber.gro -to amber -p topol.top
I am using the last version of Initram and Backward that I downloaded from Github.
In my first CG simulation, everything seems fine, I have 165 residues in both chains.
The resulting backmapped aa structure (aa_amber.gro) also seems fine with 165 residues on both chains. Even when I put it as an input to Martinize, the output of this command line is:
martinize.py -f aa_amber.gro -o topol.top -x CG.pdb -dssp /home/dssp/dsspcmbi -p backbone -ff elnedyn22
INFO Found 2 chains:
INFO 1: A (Protein), 2581 atoms in 165 residues.
INFO 2: B (Protein), 2581 atoms in 165 residues.
but then the resulting CG structure is missing residues. And the Itp files of (only) chain B misses 2 residues.
The command line of backward is:
./initram.sh -f last_frame.gro -o aa_amber.gro -to amber -p topol.top
I am using the last version of Initram and Backward that I downloaded from Github.
Please Log in or Create an account to join the conversation.
- z.sichaib
-
Topic Author
- Offline
- Senior Boarder
Less
More
- Posts: 43
4 years 1 month ago - 4 years 1 month ago #8457
by z.sichaib
Replied by z.sichaib on topic Martinize.py - missing residues in the created itp files
I also put the last frame of my unbiased CG trajectory ( last_frame.gro) here:
www.dropbox.com/sh/fb5imf64anytj60/AABoC...k0wCj59WXVgLNDa?dl=0
Last edit: 4 years 1 month ago by z.sichaib.
Please Log in or Create an account to join the conversation.
- peterkroon
-
- Offline
- Gold Boarder
Less
More
- Posts: 210
4 years 1 month ago #8458
by peterkroon
Replied by peterkroon on topic Martinize.py - missing residues in the created itp files
Ok, interesting.
Just to humor me, could you try running martinize on aa_amber.gro (freshly downloaded from your link) again, and posting /all/ the output and the exact command line? And check that it's still showing the issue. My guess is we're looking at wrong/different/old files. So just to be sure :)
Next question, which residues are missing?
Just to humor me, could you try running martinize on aa_amber.gro (freshly downloaded from your link) again, and posting /all/ the output and the exact command line? And check that it's still showing the issue. My guess is we're looking at wrong/different/old files. So just to be sure :)
Next question, which residues are missing?
Please Log in or Create an account to join the conversation.
- z.sichaib
-
Topic Author
- Offline
- Senior Boarder
Less
More
- Posts: 43
4 years 1 month ago - 4 years 1 month ago #8459
by z.sichaib
Replied by z.sichaib on topic Martinize.py - missing residues in the created itp files
Of course! thanks a lot for your help :)
martinize.py -f aa_amber.gro -o topol.top -x CG.pdb -dssp /home/dssp/dsspcmbi -p backbone -ff elnedyn22
output
INFO MARTINIZE, script version 2.6
INFO If you use this script please cite:
INFO de Jong et al., J. Chem. Theory Comput., 2013, DOI:10.1021/ct300646g
INFO Chain termini will be charged
INFO Residues at chain brakes will not be charged
INFO The elnedyn22 forcefield will be used.
INFO Local elastic bonds will be used for extended regions.
INFO Position restraints will be generated.
WARNING Position restraints are only enabled if -DPOSRES is set in the MDP file
INFO Read input structure from file.
INFO Input structure is a GRO file. Chains will be labeled consecutively.
INFO Found 2 chains:
INFO 1: A (Protein), 2581 atoms in 165 residues.
INFO 2: B (Protein), 2581 atoms in 165 residues.
INFO Total size of the system: 330 residues.
==== Secondary Structure Definition by the program DSSP, updated CMBI version by ElmK / April 1,2010 ====
==== Secondary Structure Definition by the program DSSP, updated CMBI version by ElmK / April 1,2010 ====
INFO Writing coarse grained structure.
INFO (Average) Secondary structure has been determined (see head of .itp-file).
INFO Created coarse-grained topology
INFO Created coarse-grained topology
INFO Written 2 ITP files
INFO Output contains 2 molecules:
INFO 1-> Protein_A (chain A)
INFO 2-> Protein_B (chain B)
INFO Written topology files
INFO The elnedyn forcefield has been implemented with some changes compared to the published parameters:
INFO - Backbone-Backbone bonds use high force constant bonds instead of constraints.
INFO - Trp has an extra constrain added to the sidechain.
INFO - The Backbone sidechain bonds with high force constants are replaced by constraints except for Trp and His.
INFO - Cysteine bonds are 0.24 nm constraints, instead of the published 0.39nm/5000kJ/mol.
WARNING Elnedyn topologies might not give numerical stable simulations with a 20fs timestep.
WARNING This can be solved by setting all S-type bead masses to 72amu.
The generated files from this command line are CG.pdb, Protein_B.itp, topol.top, Protein_A.itp, chain_A.ssd, chain_B.ssd. You can find these files from my link.
I did other tests and I think the problem is with DSSP.
When I run the DSSP program on my target:
dssp aa_amber.pdb out.dssp
output
!!! Backbone incomplete for residue SER 165 A
residue will be ignored !!!
!!! Backbone incomplete for residue SER 165 B
residue will be ignored !!!
!!! 2552 hydrogen or deuterium atoms were ignored
in the calculation of side chain solvent
accessibility !!!
!!! HEADER-card missing !!!
!!! COMPOUND-card missing !!!
!!! SOURCE-card missing !!!
!!! AUTHOR-card missing !!!
This might explain the problem: DSSP is ignoring the last residues and therefore martinize can't assign any bead type if the residue is already missing in the *.ssd
But, I still can't understand why DSSP outputs the backbone is incomplete for res SER 165. I checked again with the atom naming of Cter SER (amber FF) and no atom is missing!
Would you please try to run DSSP and martinize with my files and see if you can reproduce the errors? or maybe the problem is with my DSSP version? although I used before and I never had such problems
To answer your questions, the residues that are missing are GLY 164 and SER 165 (chain B).
Files are here: www.dropbox.com/sh/fb5imf64anytj60/AABoC...k0wCj59WXVgLNDa?dl=0
Thanks a lot!
martinize.py -f aa_amber.gro -o topol.top -x CG.pdb -dssp /home/dssp/dsspcmbi -p backbone -ff elnedyn22
output
INFO MARTINIZE, script version 2.6
INFO If you use this script please cite:
INFO de Jong et al., J. Chem. Theory Comput., 2013, DOI:10.1021/ct300646g
INFO Chain termini will be charged
INFO Residues at chain brakes will not be charged
INFO The elnedyn22 forcefield will be used.
INFO Local elastic bonds will be used for extended regions.
INFO Position restraints will be generated.
WARNING Position restraints are only enabled if -DPOSRES is set in the MDP file
INFO Read input structure from file.
INFO Input structure is a GRO file. Chains will be labeled consecutively.
INFO Found 2 chains:
INFO 1: A (Protein), 2581 atoms in 165 residues.
INFO 2: B (Protein), 2581 atoms in 165 residues.
INFO Total size of the system: 330 residues.
==== Secondary Structure Definition by the program DSSP, updated CMBI version by ElmK / April 1,2010 ====
==== Secondary Structure Definition by the program DSSP, updated CMBI version by ElmK / April 1,2010 ====
INFO Writing coarse grained structure.
INFO (Average) Secondary structure has been determined (see head of .itp-file).
INFO Created coarse-grained topology
INFO Created coarse-grained topology
INFO Written 2 ITP files
INFO Output contains 2 molecules:
INFO 1-> Protein_A (chain A)
INFO 2-> Protein_B (chain B)
INFO Written topology files
INFO The elnedyn forcefield has been implemented with some changes compared to the published parameters:
INFO - Backbone-Backbone bonds use high force constant bonds instead of constraints.
INFO - Trp has an extra constrain added to the sidechain.
INFO - The Backbone sidechain bonds with high force constants are replaced by constraints except for Trp and His.
INFO - Cysteine bonds are 0.24 nm constraints, instead of the published 0.39nm/5000kJ/mol.
WARNING Elnedyn topologies might not give numerical stable simulations with a 20fs timestep.
WARNING This can be solved by setting all S-type bead masses to 72amu.
The generated files from this command line are CG.pdb, Protein_B.itp, topol.top, Protein_A.itp, chain_A.ssd, chain_B.ssd. You can find these files from my link.
I did other tests and I think the problem is with DSSP.
When I run the DSSP program on my target:
dssp aa_amber.pdb out.dssp
output
!!! Backbone incomplete for residue SER 165 A
residue will be ignored !!!
!!! Backbone incomplete for residue SER 165 B
residue will be ignored !!!
!!! 2552 hydrogen or deuterium atoms were ignored
in the calculation of side chain solvent
accessibility !!!
!!! HEADER-card missing !!!
!!! COMPOUND-card missing !!!
!!! SOURCE-card missing !!!
!!! AUTHOR-card missing !!!
This might explain the problem: DSSP is ignoring the last residues and therefore martinize can't assign any bead type if the residue is already missing in the *.ssd
But, I still can't understand why DSSP outputs the backbone is incomplete for res SER 165. I checked again with the atom naming of Cter SER (amber FF) and no atom is missing!
Would you please try to run DSSP and martinize with my files and see if you can reproduce the errors? or maybe the problem is with my DSSP version? although I used before and I never had such problems
To answer your questions, the residues that are missing are GLY 164 and SER 165 (chain B).
Files are here: www.dropbox.com/sh/fb5imf64anytj60/AABoC...k0wCj59WXVgLNDa?dl=0
Thanks a lot!
Last edit: 4 years 1 month ago by z.sichaib.
Please Log in or Create an account to join the conversation.
- peterkroon
-
- Offline
- Gold Boarder
Less
More
- Posts: 210
4 years 1 month ago #8461
by peterkroon
Replied by peterkroon on topic Martinize.py - missing residues in the created itp files
I think I know what's going on :)
Your C-terminal oxygens are called OC1 and OC2. DSSP does not like/recognize these, and in response throws away *both* C-terminal residues. Martinize then gets confused and simply keep processing residues as long has it has DSSP output to work with. So all secondary structure assignments for chain B are shifted by one residue.
Rename your OC1 and OC2 to O and OXT and everything works as expected :)
Your C-terminal oxygens are called OC1 and OC2. DSSP does not like/recognize these, and in response throws away *both* C-terminal residues. Martinize then gets confused and simply keep processing residues as long has it has DSSP output to work with. So all secondary structure assignments for chain B are shifted by one residue.
Rename your OC1 and OC2 to O and OXT and everything works as expected :)
Please Log in or Create an account to join the conversation.
- z.sichaib
-
Topic Author
- Offline
- Senior Boarder
Less
More
- Posts: 43
4 years 1 month ago #8462
by z.sichaib
Replied by z.sichaib on topic Martinize.py - missing residues in the created itp files
It's working! thanks a lot :D
Please Log in or Create an account to join the conversation.
Time to create page: 0.113 seconds