Adrian Schreyer

CREDO version 01.2010 released!

A new version of the CREDO database has been released today comprising 41,691 protein-ligand complexes from the PDB.

New features include:

  • All contacts up to 6A are now stored
  • A new Variations tables has been added containing mutation information from dbSNP, EnsEMBL, OMIM and COSMIC
  • The Ligands table now contains an ism field to store isomeric SMILES (very useful for heteropeptides)
  • The LigandRing and ResidueRing tables now contain an is_hetero_aromatic field
  • FragmentProperties now contains all descriptors from OEMolProp
  • Several improvements to the CREDO API
    • For instructions on how to download and install the CREDO database, please go to the CREDO website.

Identifying novel protein-protein-interaction inhibitors in CREDO

Identifying novel protein-protein-interaction inhibitors in CREDO

This PyMOL screenshot shows the alignment of two PDB structures, the first, a caspase-3 complexed with a nonpeptidic inhibitor (molecule shown as sticks) and the second, caspase-3 in complex with XIAP-BIR2 (shown as cartoon).

VEGFR2 in complex with a novel 4-amino-furo[2,3-d]pyrimidine inhibitor (SNPs and modified residues highlighted)

VEGFR2 in complex with a novel 4-amino-furo[2,3-d]pyrimidine inhibitor (SNPs and modified residues highlighted)

In [1]: from credopymol import *
In [2]: s = StructureAdaptor().fetchByPDB('1YWN')
In [3]: s.load() # modified residues are automatically highlighted in green
 
In [4]: s.Ligands 
Out[4]: [Ligand(301, LIF, A)]
 
In [5]: s.Ligands[0].showContacts()
In [6]: s.showMutations()
 
In [7]: s.getAbstract ()

OMIM mutations highlighted in insulin receptor crystal structure

OMIM mutations highlighted in insulin receptor crystal structure

In [1]: from credopymol import *
In [2]: s = StructureAdaptor().fetchByPDB('3BU5')
 
In [3]: s
Out[3]: <Structure('3BU5')>
 
In [4]: s.title
Out[4]: 'CRYSTAL STRUCTURE OF THE INSULIN RECEPTOR KINASE IN COMPLEX WITH IRS2 KRLB PEPTIDE AND ATP'
 
In [5]: s.load()
 
In [5]: s.getMutations()
Out[5]:
[<XRef(Residue,OMIM,610549)>,
 <XRef(Residue,OMIM,125853)>,
 <XRef(Residue,OMIM,125853)>]
 
In [6]: mutations = s.getMutations()
 
In [7]: mutations[0].description
Out[7]: 'G/V Insulin-resistant diabetes mellitus with acanthosis nigricans type A (IRAN type A)'

Natural variant in 'Warrior Gene' amine oxidase A

Natural variant in 'Warrior Gene' amine oxidase A

In [1]: from credopymol import *
In [2]: s = StructureAdaptor().fetchByPDB('2Z5X')
In [3]: s.load()
 
In [4]: xref = s.getMutations()[0]
 
In [5]: xref
Out[5]: <XRef(Residue,dbSNP,rs1803986)>
 
In [6]: xref.description 
Out[6]: 'M/I NON_SYNONYMOUS_CODING'
 
In [7]: mut_res = xref.getObject()
 
In [8]: mut_res
Out[8]: <Residue(445, MET, )>
 
In [9]: mut_res.show()
 
In [10]: s.Ligands[3] 
Out[10]: Ligand(600, FAD, A)
 
In [11]: s.Ligands[3].showContacts()

Aligned CDK2 / CDK4 mimic (CDK2 mutant) structures in complex with the same inhibitor and mutations highlighted

Aligned CDK2 / CDK4 mimic (CDK2 mutant) structures in complex with the same inhibitor and mutations highlighted

The screenshot shows the binding site of 1GIH and 1GII with the mutated residues in the latter highlighted in magenta (carbon atoms).

Alignment of CDK2 and JNK crystal structures complexed with inhibitors

Alignment of CDK2 and JNK crystal structures complexed with inhibitors

The image shows the structural interactions of the bound inhibitors in the aligned chains 'A' of 2C4G and 2P33. Both ligands were found to be leaves of the same node in a SCOP-based SIFt clustering.

Crystal structures of human CDK2 complexed with inhibitor

Crystal structures of human CDK2 complexed with inhibitor

The image shows two crystal structures of human CDK2 in complex with inhibitors (1Y8Y, 2C6L). Structural interaction fingerprints (SIFts) were clustered (based on UniProt accession of the protein) and these two ligands were found to be neighbouring leaves on the same node. The structures can be loaded into PyMOL, aligned and the interactions visualised with command below. The resulting PyMOL scene can be downloaded from the bottom of this page.

Fragment Contact Density

Fragment Contact Density

In [1]: from credopymol import *
In [2]: s = StructureAdaptor().fetchByPDB('2P33')
In [3]: s.load()
In [4]: s.Ligands[0].showContacts()
In [5]: s.Ligands[0].showFragmentContactDensity()

Syndicate content