this error is related to the phosphorilated residues (SER, THR) appearing in your structure. These residues are not in the standard force field (in this case Amber99SB*). It is not an easy problem to solve for an automatic platform like MDWeb.
First of all I’m sorry to tell you that MDWeb is not able to work with modified residues (covalently bonded residues) in GROMACS. You can do it in AMBER though, but it requires a number of steps:
You should rename these residues in the PDB file to distinguish them from the standard ones: from SER to SEP and from THR to THP. You should also change the ATOM tag for the HETATM tag. Then MDWeb will be able to recognise them as ligands, and will offer you the possibility to use the parameters coming from the Manchester AMBER parameter database: http://research.bmh.manchester.ac.uk/bryce/amber
One more thing is to rename the atom names in the PDB file to match those in the parameter files (P01 to PS, O01 to OPA, O02 to OPB and O03 to OPC, please see https://personalpages.manchester.ac.uk/staff/Richard.Bryce/amber/pro/amber_sep.lib) or just use the ligand checking utility from MDWeb (http://mmb.irbbarcelona.org/MDWeb/help.php?id=checkingLig) to match them.
And still one last thing. I think your structure has a phosphorylated serine in the N-terminal. As the parameters available are for the phospho-serine (not N-terminal phospho-serine), the structure will not be properly prepared with MDWeb. Instead, you can just add a new capping residue or just remove the first phospho-serine (if you know it is not crucial for the protein).
Hope this helps! Thanks for using our tool!