Discovering Drug-Protein Interactions Based on their Fingerprints

The unveiling of rules that govern drug-protein interactions is of paramount importance in drug discovery. To discover such relationships, we propose to use a novel method called DPA. Given a set of drug-protein interactions, DPA performs its tasks in several steps: (i) for each drug involved, its substructures are each converted into its fingerprints, (ii) for each protein involved, its protein domains are each converted into its fingerprints, (iii) a dependency measure between each drug substructure and protein domain is then computed based on the known interactions between the drugs and proteins, (iv) the dependency measures are then used to predict previously unknown drug-protein interactions. DPA has the advantage that it is able to perform its tasks effectively without requiring any 3D information about drug and protein structures. It makes use of molecular fingerprints which are information-rich and fast to compute. DPA has been tested with known drug-protein interaction data including enzymes, ion channels, protein-coupled receptors. Experimental results show that it can be very useful for predicting new drug-protein interaction as well as protein-ligand interactions. It can also be used to tackle problems such as ligand specificity thereby facilitating the drug discovery process.

[1]  Z. Deng,et al.  Structural interaction fingerprint (SIFt): a novel method for analyzing three-dimensional protein-ligand binding interactions. , 2004, Journal of medicinal chemistry.

[2]  Yoshihiro Yamanishi,et al.  Extracting Sets of Chemical Substructures and Protein Domains Governing Drug-Target Interactions , 2011, J. Chem. Inf. Model..

[3]  Marvin Johnson,et al.  Concepts and applications of molecular similarity , 1990 .

[4]  Michael Gribskov,et al.  Use of Receiver Operating Characteristic (ROC) Analysis to Evaluate Sequence Matching , 1996, Comput. Chem..

[5]  R. Desimone,et al.  Privileged structures: applications in drug discovery. , 2004, Combinatorial chemistry & high throughput screening.

[6]  Malcolm J. McGregor,et al.  Clustering of Large Databases of Compounds: Using the MDL "Keys" as Structural Descriptors , 1997, J. Chem. Inf. Comput. Sci..

[7]  Michael J. Keiser,et al.  Predicting new molecular targets for known drugs , 2009, Nature.

[8]  Andrew K. C. Wong,et al.  Learning sequential patterns for probabilistic inductive prediction , 1994 .

[9]  Frederick P. Roth,et al.  Chemical substructures that enrich for biological activity , 2008, Bioinform..

[10]  David M. Rocke,et al.  Predicting ligand binding to proteins by affinity fingerprinting. , 1995, Chemistry & biology.

[11]  Daniel C. Harris,et al.  Quantitative Chemical Analysis , 1968, Nature.

[12]  Bin Chen,et al.  PubChem as a Source of Polypharmacology , 2009, J. Chem. Inf. Model..

[13]  Jürgen Bajorath,et al.  Design and Evaluation of a Molecular Fingerprint Involving the Transformation of Property Descriptor Values into a Binary Classification Scheme , 2003, J. Chem. Inf. Comput. Sci..

[14]  Baris E. Suzek,et al.  The Universal Protein Resource (UniProt) in 2010 , 2009, Nucleic Acids Res..

[15]  Pedro Alexandrino Fernandes,et al.  Protein–ligand docking: Current status and future challenges , 2006, Proteins.

[16]  María Martín,et al.  The Universal Protein Resource (UniProt) in 2010 , 2010 .

[17]  S. Haberman The Analysis of Residuals in Cross-Classified Tables , 1973 .

[18]  L. Holm,et al.  The Pfam protein families database , 2005, Nucleic Acids Res..

[19]  Michael J. Keiser,et al.  Relating protein pharmacology by ligand chemistry , 2007, Nature Biotechnology.

[20]  Robert D. Finn,et al.  The Pfam protein families database , 2004, Nucleic Acids Res..