1.Databases and web servers
(1)DrugBank(http://www.drugbank.ca)

(2)TTD:Therapeutic target database(http://bidd.nus.edu.sg/group/ttd/ttd.asp)

(3)SuperTarget(http://bioinf-apache.charite.de/supertarget_v2/)

(4)MATADOR(http://matador.embl.de)

(5)STITCH(http://stitch.embl.de/)

(6)TDR targets(http://tdrtargets.org/)
 (7)PDTD(http://www.dddc.ac.cn/pdtd/)

 (8)ChEMBL(https://www.ebi.ac.uk/chembldb)

(9)Integrity(http://integrity.thomson-pharma.com)
 (10)FAERS(http://www.fda.gov/Drugs/GuidanceComplianceRegulatoryInformation/Surveillance/AdverseDrugEffects/default.htm) 

(11)SIDER(http://sideeffects.embl.de/) 

(12)JAPIC(http://www.japic.or.jp/) 

(13)ChemBank(http://chembank.broadinstitute.org/) 

(14)The IUPHAR/BPS Guide to PHARMACOLOGY(http://www.guidetopharmacology.org/) 

(15)CancerDR(http://crdd.osdd.net/raghava/cancerdr/) 

(16)BindingDB(http://www.bindingdb.org/bind) 

(17)ZINC(http://zinc.docking.org) 

(18)canSAR(https://cansar.icr.ac.uk) 

(19)PDSP(http://pdsp.med.unc.edu)
(20)DCDB(http://www.cls.zju.edu.cn/dcdb/)  

(21)ASDCD(http://asdcd.amss.ac.cn/) 

(22)DINIES(http://www.genome.jp/tools/dinies/) 

(23)SuperPred(http://prediction.charite.de/) 

(24)SwissTargetPrediction(http://www.swisstargetprediction.ch/)

 
Table.The statistics of the number of drugs, target proteins and drug–target interactions in some of the databases

2.Computational models

(1)Network-based model

a.MTOI

(2)Drug side-effect similarity-based method
a.NRWRH

b.DBSI, TBSI and NBI

c.Within scores and between scores

(3)Machine learning-based method

a.Supervised learning method

b.Bipartite graph learning method

c.BLM

d.Bipartite graph learning method by introducing the pharmacological data

e.BLM-NII

f.RBM

g.Random forest

h.Two strategies for negative sample selection

i.Semi-supervised learning method

j.NetLapRLS

k.Kron-RLS

(4)Other methods

a.Chemical similarities

b.A two-step similarity-based method

c.PMF

Reference:
-----------------------
1.Drug–target interaction prediction: databases, web servers and computational models.