Modified amino acids problem for the conservation pipeline

[kiarka7]

The conservation pipeline falls down when it runs on modified amino acids. Currently, DODO cannot process modified amino acids. Although the .pdbqt is generated and the docking is done, for the conservation pipeline the structure is incomplete - the protein part after the modified amino acid is missing. Also the modified amino acid itself is not part of the protein. See 3LTH structure statement:

RuntimeError: Sequences for chain A region (0, 419) expected: 'MDKFRVQGPTRLQGEVTISGAKNAALPILFAALLAEEPVEIQNVPKLKDIDTTMKLLTQLGTKVERDGSVWIDASNVNNFSAPYDLVKTMRASIWALGPLVARFGQGQVSLPGGCAIGARPVDLHIFGLEKLGAEIKLEEGYVKASVNGRLKGAHIVMDKVSVGATVTIMSAATLAEGTTIIENAAREPEIVDTANFLVALGAKISGQGTDRITIEGVERLGGGVYRVLPDRIETGTFLVAAAISGGKIVCRNAQPDTLDAVLAKLREAGADIETGEDWISLDMHGKRPKAVTVRTAPHPAFPTDMQAQFTLLNLVAEGTGVITETIFENRFMHVPELIRMGAHAEIESNTVICHGVEKLSGAQVMATDLRASASLVLAGCIAEGTTVVDRIYHIDRGYERIEDKLRALGANIERVKGE' actual: 'MDKFRVQGPTRLQGEVTISGAKNAALPILFAALLAEEPVEIQNVPKLKDIDTTMKLLTQLGTKVERX' must have same size (419, 67).

In the 67th position is originally IAS, a non-standard amino acid: BETA-L-ASPARTIC ACID.

[rdk]

Thanks for reporting this.

This is caused by P2Rank bug that cuts chains after a residue containing OXT atom.

In the pdb file there is HETATM 532 OXT IAS A 67 and the problem is that the atom type is OXT, not that the residue is modified per se. As a workaround it should be sufficient to manually change it to O.

All versions of P2Rank up to 2.4.1 are affected. The bug has been fixed on the develop branch and we are preparing the release.

[luk27official]

P2Rank 2.4.2 has been released, so I think we might upgrade the version to fix this bug. :)
https://github.com/rdk/p2rank/releases/tag/2.4.2