genomes such as human and mouse are linked to higher-level functions of the cell

and the individual organisms (Kanehisa and Goto 2000). It contains a collection of

manually drawn pathway maps representing knowledge on the biochemical

pathways, pathways of genetic information processing, pathways of environmental

information processing, organismal systems, human genetic disorders, and drug

development. The pathway database, BioCyc, consists of predicted pathway infor-

mation from model organisms and provides various tools for pathway analysis (Karp

et al. 2005). Reactome contains a manually curated open-source data of human

pathways, reactions, bioinformatics analysis, and visualization tools (Croft et al.

2014). The Molecular Signatures Database (MSigDB) contains gene-sets belonging

to particular pathways, chromosomal locations, protein complexes, and transcription

factor binding (Liberzon et al. 2011).

4.2.3

Drug-Target Interactions

The database DrugBank is a unique genomics and cheminformatics resource that

integrates drug and nutraceuticals data with all-known drug target information. The

recent version of DrugBank (version 5.1.6) contains information on 13,579 drugs

that consists of 2635 approved small molecule drugs, 131 nutraceuticals, and over

6375 drugs in discovery phase (Wishart et al. 2018). The database BindingDB is a

publicly available repository of experimentally measured binding affinities related to

interactions between small, drug-like molecules and drug-targets. BindingDB

contains 1,881,721 binding data, for 7548 protein targets and 833,792 small

molecules (Liu et al. 2007). The Drug-Gene Interaction database (DGIdb) mines

existing data from 30 disparate sources that provides information about how mutated

genes could be targeted therapeutically or prioritized for drug development. The

database also provides a web interface for searching a set of genes against a

collection of drug-gene interactions and potential druggable genes (Griffith et al.

2013). The Pharmacogenomics Knowledge Base (PharmGKB) consists of the data-

base on the effect of various genetic variations on drug response (Hewett et al. 2002).

The database ChEMBL is a repository of bioactive drug-like small molecules,

two-dimensional

structures,

computed

physicochemical

properties,

and

bio-activities such as binding constants, pharmacology, and ADMET information

(Bento et al. 2014). Therapeutic Target Database (TTD) is a repository that provides

information about the known and predicted protein and nucleic acid targets, drugs

directed at each of the targets, and pathway information on the targets. In addition,

the database also consists of links to other relevant databases containing information

about target function, ligand binding properties, enzyme nomenclature and drug

structure, therapeutic class, clinical development status, target sequence, and 3D

structure (Chen et al. 2002). Another database, search tool for interactions of

chemicals (STITCH), integrates information about crystal structures, binding

experiments, drug-target relationships, and interactions from metabolic pathways

(Kuhn et al. 2007). STITCH also contains the network of chemical relations

depending on chemical similarity and information for over 70,000 different

4

Computational Methods for Drug Repurposing

41