OpenMM Residue Value Error

[mottopanikeiku]

Describe the bug
What happened: BP.generate_systems() causes a ValueError with openMM
ValueError: No template found for residue 357 (POP). The set of atoms matches POPC, but the bonds are different. Perhaps the chain is missing a terminal group? For more information, see https://github.com/openmm/openmm/wiki/Frequently-Asked-Questions#template

Expected behavior
What did you expect to happen:
The forcefield to be built
To Reproduce
Steps to reproduce the behavior:
1.Sequence was changed from RCSB Fasta sequence to the correct sequence
2.Code was ran normally.

The issue was resolved when the secondary_template was changed to None.

Screenshots
If applicable, add screenshots to help explain your problem.

6DRY.pdb
6dry_opm.pdb
6DRY_WT_SECONDARY.ipynb

5HT2B_AF.pdb

The notebook attached above might not contain the correct sequence, if that is the case, the new sequence attempted was
MALSYRVSELQSTIPEHILQSTFVHVISSNWSGLQTESIPEEMKQIVEEQGNKLHWAALLILMVIIPTIGGNTLVILAVSLEKKLQYATNYFLMSLAVADLLVGLFVMPIALLTIMFEAMWPLPLVLCPAWLFLDVLFSTASIMHLCAISVDRYIAIKKPIQANQYNSRATAFIKITVVWLISIGIAIPVPIKGIETDVDNPNNITCVLTKERFGDFMLFGSLAAFFTPLAIMIVTYFLTIHALQKKAYLVKNKPPQRLTWLTVSTVFQRDETPCSSPEKVAMLDGSRKDKALPNSGDETLMRRTSTIGKKSVQTISNEQRASKVLGIVFFLFLLMWCPFFITNITLVLCDSCNQTTLQMLLEIFVWIGYVSSGVNPLVYTLFNKTFRDAFGRYITCNYRATKSVKTLRKRSSKIYFRNPMAENSKFFKKHGIRNGINPAMYQSPMRLRSSTIQSSSIILLDTLLLTENEGDKTEEQVSYV

[aimankhan131]

Hey, I found the issue. There's a bug in Bridgeport.py around line 541 that tries to renumber residues after adding the membrane. This renumbering accidentally breaks the lipid bond structures, which is why OpenMM can't recognize them as POPC. I just commented out that section, so it skips the renumbering. Now the lipids stay intact, and everything works fine. And I used a secondary template and I used this sequence:

MALSYRVSELQSTIPEHILQSTFVHVISSNWSGLQTESIPEEMKQIVEEQGNKLHWAALLILMVIIPTIGGNTLVILAVSLEKKLQYATNYFLMSLAVADLLVGLFVMPIALLTIMFEAMWPLPLVLCPAWLFLDVLFSTASIMHLCAISVDRYIAIKKPIQANQYNSRATAFIKITVVWLISIGIAIPVPIKGIETDVDNPNNITCVLTKERFGDFMLFGSLAAFFTPLAIMIVTYFLTIHALQKKAYLVKNKPPQRLTWLTVSTVFQRDETPCSSPEKVAMLDGSRKDKALPNSGDETLMRRTSTIGKKSVQTISNEQRASKVLGIVFFLFLLMWCPFFITNITLVLCDSCNQTTLQMLLEIFVWIGYVSSGVNPLVYTLFNKTFRDAFGRYITCNYRATKSVKTLRKRSSKIYFRNPMAENSKFFKKHGIRNGINPAMYQSPMRLRSSTIQSSSIILLDTLLLTENEGDKTEEQVSYV.
I've attached json file and pdb file that I got after completed the Bridgeport job for your reference.

methylergonovine.pdb

6DRY_WT_SECONDARY_fixed(1).ipynb

Thankyou.

Regards
Aiman Khan

[davidcooper7]

I'm skeptical that the following code alters the bond in the topology (from https://github.com/CCBatIIT/Bridgeport/blob/main/Bridgeport/Bridgeport.py#L536)

        # Correct resids
        pdb = md.load_pdb(self.env_pdb)
        top = pdb.topology
        if top.residue(0).resSeq == 1:
            for i in range(top.n_residues):
                top.residue(i).resSeq += self.tails[0]
            pdb[0].save_pdb(self.env_pdb)

But if that happens to be the case, I think a better fix would be:

        # Correct resids
        pdb = md.load_pdb(self.env_pdb)
        top = pdb.topology
        if top.residue(0).resSeq == 1:
            for i in range(top.n_residues):
                if top.residue(i).name not in ['POP', 'HOH', 'NA', 'CL']: #Only renumber protein residues
                    top.residue(i).resSeq += self.tails[0]
            pdb[0].save_pdb(self.env_pdb)

Can you try this instead of just commenting this out? Renumbering the residues of the protein is critical for analysis and consistency amongst molecular dynamics simulations.

[daveminh]

Maybe better to explicitly list protein residue names in case we later use different lipids? David Minh ***@***.******@***.***> From: davidcooper7 ***@***.***> Date: Wednesday, December 10, 2025 at 11:45 PM To: CCBatIIT/Bridgeport ***@***.***> Cc: Subscribed ***@***.***> Subject: Re: [CCBatIIT/Bridgeport] OpenMM Residue Value Error (Issue #43) [https://avatars.githubusercontent.com/u/117563200?s=20&v=4]davidcooper7 left a comment (CCBatIIT/Bridgeport#43)<#43 (comment)> I'm skeptical that the following code alters the bond in the topology (from https://github.com/CCBatIIT/Bridgeport/blob/main/Bridgeport/Bridgeport.py#L536) # Correct resids pdb = md.load_pdb(self.env_pdb) top = pdb.topology if top.residue(0).resSeq == 1: for i in range(top.n_residues): top.residue(i).resSeq += self.tails[0] pdb[0].save_pdb(self.env_pdb) I think a better fix would be # Correct resids pdb = md.load_pdb(self.env_pdb) top = pdb.topology if top.residue(0).resSeq == 1: for i in range(top.n_residues): if top.residue(i).name not in ['POP', 'HOH', 'NA', 'CL']: #Only renumber protein residues top.residue(i).resSeq += self.tails[0] pdb[0].save_pdb(self.env_pdb) — Reply to this email directly, view it on GitHub<#43 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ABQDIOD72Q6V23H67TNHYR34BEAG7AVCNFSM6AAAAACMBNUEGOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTMNBQGI3DIMJZGY>. You are receiving this because you are subscribed to this thread.

[davidcooper7]

This commit 5b1ad63 should fix the issue. I substituted mdtraj for openmm topology renumbering, which preserves CONECT records. This appears to fix the issue.

@mottopanikeiku please update your code base and try.