In this section we illustrate the use of Netpredictor package in prediction of drug target interactions and analysis of networks. The information about the interactions between drugs and target proteins was obtained from Yamanishi etal \cite{bib33} where the number of drugs 212, 99, 105 and 27, interacting with enzymes, ion channels, GPCRs and nuclear receptors respectively. The numbers of the corresponding target proteins in these classes are 478, 146, 84 and 22 respectively. The numbers of the corresponding interactions are 1515, 776, 314 and 44. We performed both network based inference and Random walk with restart on all of these datasets. To check the performance we randomly removed 20\% of the interactions from each of the dataset and computed the performance 50 times and calculated the mean performance of each of these methods . The results are given in Table 3. Clearly, RWR supersedes its performance compared to network based inference in Enzyme and the GPCR dataset. However, computation of NBI algorithm takes less amount of time than RWR. For the drugbank tab we download the latest drugbank set version 4.3 and created a drug target interaction list of 5970 drugs and 3797 proteins We computed similarities of drugs using RDkit\cite{bib34} ECFP6 fingerprint and local sequence similarity of proteins using smith waterman algorithm and normalized using the procedure proposed by Bleakley and Yamanishi\cite{bib35} and integrated the matrices for network based inference computation. We ran the computations 50 times and kept the significant drug target relations(p $\leqslant$ 0.05) where a total of  316645 predicted interactions and 14167 true interactions present in the system.