DINIES: drug–target interaction network inference engine based on supervised analysis

In this work, we performed an analysis of the networks of interactions between drugs and their targets to assess how connected the compounds are. For our purpose, the interactions were downloaded from the DrugBank database, and we considered all drugs approved by the FDA. Based on topological analysis of this interaction network, we obtained information on degree, clustering coefficient, connected components, and centrality of these interactions. We identified that this drug-target interaction network cannot be divided into two disjoint and independent sets, i.e., it is not bipartite. In addition, the connectivity or associations between every pair of nodes identified that the drug-target network is constituted of 165 connected components, where one giant component contains 4376 interactions that represent 89.99% of all the elements. In this regard, the histamine H1 receptor, which belongs to the family of rhodopsin-like G-protein-coupled receptors and is activated by the biogenic amine histamine, was found to be the most important node in the centrality of input-degrees. In the case of centrality of output-degrees, fostamatinib was found to be the most important node, as this drug interacts with 300 different targets, including arachidonate 5-lipoxygenase or ALOX5, expressed on cells primarily involved in regulation of immune responses. The top 10 hubs interacted with 33% of the target genes. Fostamatinib stands out because it is used for the treatment of chronic immune thrombocytopenia in adults. Finally, 187 highly connected sets of nodes, structured in communities, were also identified. Indeed, the largest communities have more than 400 elements and are related to metabolic diseases, psychiatric disorders and cancer. Our results demonstrate the possibilities to explore these compounds and their targets to improve drug repositioning and contend against emergent diseases.

Download Full-text

Drug-target Interaction Prediction by Metapath2vec Node Embedding in Heterogeneous Network of Interactions

International Journal of Artificial Intelligence Tools ◽

10.1142/s0218213020500013 ◽

2020 ◽

Vol 29 (01) ◽

pp. 2050001

Author(s):

Mina Samizadeh ◽

Behrouz Minaei-Bidgoli

Keyword(s):

Drug Development ◽

Heterogeneous Network ◽

Drug Target ◽

Network Inference ◽

Interaction Network ◽

Ensemble Methods ◽

New Drugs ◽

Molecular Entity ◽

Heterogeneous Interaction ◽

Scale Down

Drug discovery is a complicated, time-consuming and expensive process. The cost for each new molecular entity (NME) is estimated at $1.8 billion. Furthermore, for a new drug to be FDA approved it often takes nearly a decade and approximately 20 new drugs being approved by the US Food and Drug Administration (FDA) each year. Accurately predicting drug-target interactions (DTIs) by computational methods is an important area of drug research, which brings about a broad prospect for fast and low-risk drug development. By accurate prediction of drugs and targets interactions scientists can scale-down huge experimental space and reduce the costs and help to faster drug development as well as predicting the side effects and potential function of new drugs. Many approaches have been taken by researchers to solve DTI problem and enhance the accuracy of methods. State-of-the-art approaches are based on various techniques, such as deep learning methods-like stacked auto-encoder-, matrix factorization, network inference, and ensemble methods. In this work, we have taken a new approach based on node embedding in a heterogeneous interaction network to obtain the representation of each node in the interaction network and then use a binary classifier such as logistic regression to solve this prominent problem in the pharmaceutical industry. Most introduced network-based methods use a homogeneous network of interactions as their input data whereas in the real word problem there exist other informative networks to help to enhance the prediction and by considering the homogeneous networks we lose some precious network information. Hence, in this work, we have tried to work on the heterogeneous network and have improved the accuracy of methods in comparison to baseline methods.

Download Full-text

Drug-Target Interaction Network Predictions for Drug Repurposing Using LASSO-Based Regularized Linear Classification Model

Advances in Artificial Intelligence - Lecture Notes in Computer Science ◽

10.1007/978-3-319-89656-4_26 ◽

2018 ◽

pp. 272-278 ◽

Cited By ~ 1

Author(s):

Jiaying You ◽

Md. Mohaiminul Islam ◽

Liam Grenier ◽

Qin Kuang ◽

Robert D. McLeod ◽

...

Keyword(s):

Drug Target ◽

Interaction Network ◽

Drug Repurposing ◽

Classification Model ◽

Linear Classification ◽

Target Interaction

Download Full-text

Predicting drug-target interaction network using deep learning model

Computational Biology and Chemistry ◽

10.1016/j.compbiolchem.2019.03.016 ◽

2019 ◽

Vol 80 ◽

pp. 90-101 ◽

Cited By ~ 18

Author(s):

Jiaying You ◽

Robert D. McLeod ◽

Pingzhao Hu

Keyword(s):

Deep Learning ◽

Drug Target ◽

Interaction Network ◽

Learning Model ◽

Target Interaction ◽

Deep Learning Model

Download Full-text

QPowered Compound2DeNovoDrugPropMax –A Novel Programmatic Tool Incorporating Deep Learning and In Silico Methods for Automated In Silico Bio- Activity Discovery for any Compound of Interest

10.26434/chemrxiv.13052951.v3 ◽

2021 ◽

Author(s):

Ben Geoffrey A S ◽

Rafal Madaj ◽

Akhil Sanker ◽

Pavan Preetham Valluri

Keyword(s):

Machine Learning ◽

Deep Learning ◽

In Silico ◽

Drug Target ◽

Drug Targets ◽

Learning Algorithms ◽

Interaction Network ◽

Network Data ◽

Data Set ◽

Target Interaction

Network data is composed of nodes and edges. Successful application of machine learning/deep<br>learning algorithms on network data to make node classification and link prediction have been shown<br>in the area of social networks through which highly customized suggestions are offered to social<br>network users. Similarly one can attempt the use of machine learning/deep learning algorithms on<br>biological network data to generate predictions of scientific usefulness. In the presented work,<br>compound-drug target interaction network data set from bindingDB has been used to train deep<br>learning neural network and a multi class classification has been implemented to classify PubChem<br>compound queried by the user into class labels of PBD IDs. This way target interaction prediction for<br>PubChem compounds is carried out using deep learning. The user is required to input the PubChem<br>Compound ID (CID) of the compound the user wishes to gain information about its predicted<br>biological activity and the tool outputs the RCSB PDB IDs of the predicted drug target interaction for<br>the input CID. Further the tool also optimizes the compound of interest of the user toward drug<br>likeness properties through a deep learning based structure optimization with a deep learning based<br>drug likeness optimization protocol. The tool also incorporates a feature to perform automated In<br>Silico modelling for the compounds and the predicted drug targets to uncover their protein-ligand<br>interaction profiles. The program is hosted, supported and maintained at the following GitHub<br><div>repository</div><div><br></div><div>https://github.com/bengeof/Compound2DeNovoDrugPropMax</div><div><br></div>Anticipating the rise in the use of quantum computing and quantum machine learning in drug discovery we use<br>the Penny-lane interface to quantum hardware to turn classical Keras layers used in our machine/deep<br>learning models into a quantum layer and introduce quantum layers into classical models to produce a<br>quantum-classical machine/deep learning hybrid model of our tool and the code corresponding to the<br><div>same is provided below</div><div><br></div>https://github.com/bengeof/QPoweredCompound2DeNovoDrugPropMax<br>

Download Full-text