# How to convert amino acid sequence to SMILES format

I have an amino acids string "MKGK" and I would like to convert it into SMILES format.

I tried two approaches.

First with PepSMI which gives me:

N[C@@]([H])(CCSC)C(=O)N[C@@]([H])(CCCCN)C(=O)NCC(=O)N[C@@]([H])(CCCCN)C(=O)O


Second, using ChemAxon Marvin's molconvert command

/Applications/MarvinSuite/bin/molconvert --peptide MKGK smiles


I get:

[H]N[C@@H](CCSC)C(=O)N[C@@H](CCCCN)C(=O)NCC(=O)N[C@@H](CCCCN)C(O)=O


Both of them give different result. Are they chemically the same? If not, which result should I trust?

SMILES are meant for machines, but reading up on the format is very handy and one totally can read and write them —I do it often and could not recommend it enough.

You can have SMILES written differently and they give the same result —hence why you should never compare SMILES, but Inchi or better still a parsed molecule —e.g. in RDKit.

Looking at them, I can see that

• they are the same except for one explicit proton on the N-terminus.
• Neither are canonical SMILES —which is fine.
• Both are written N to C like biologist write peptides —chemists write peptides C to N due to solid phase synthesis.
• they are L-amino acids: the pattern N[C@@H](*)C(=O)O is the give away. Technically, the implicit hydrogen in the Cα [C@@H] has @@, which means counter-clockwise —Voet & Voet taught everyone to read C(=O)RN counter-clockwise with the N-C (the first bond) sticking out.

The first has rather oddly declared chiral protons, the second has an unnecessary N-terminus proton, which is not adding a charge: [H]N* is the same as [NH2]* as it is assumed to be [H][NH]*. So the correct version at pH 7 should be:

[NH3+][C@@H](CCSC)C(=O)N[C@@H](CCCCN)C(=O)NCC(=O)N[C@@H](CCCCN)C([O-])=O