Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2023 Jun 12;24(12):10033.
doi: 10.3390/ijms241210033.

Drug Target Identification and Drug Repurposing in Psoriasis through Systems Biology Approach, DNN-Based DTI Model and Genome-Wide Microarray Data

Yu-Ping Zhan  1 , Bor-Sen Chen  1
Affiliations

Drug Target Identification and Drug Repurposing in Psoriasis through Systems Biology Approach, DNN-Based DTI Model and Genome-Wide Microarray Data

Yu-Ping Zhan et al. Int J Mol Sci. .

Abstract

Psoriasis is a chronic skin disease that affects millions of people worldwide. In 2014, psoriasis was recognized by the World Health Organization (WHO) as a serious non-communicable disease. In this study, a systems biology approach was used to investigate the underlying pathogenic mechanism of psoriasis and identify the potential drug targets for therapeutic treatment. The study involved the construction of a candidate genome-wide genetic and epigenetic network (GWGEN) through big data mining, followed by the identification of real GWGENs of psoriatic and non-psoriatic using system identification and system order detection methods. Core GWGENs were extracted from real GWGENs using the Principal Network Projection (PNP) method, and the corresponding core signaling pathways were annotated using the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways. Comparing core signaling pathways of psoriasis and non-psoriasis and their downstream cellular dysfunctions, STAT3, CEBPB, NF-κB, and FOXO1 are identified as significant biomarkers of pathogenic mechanism and considered as drug targets for the therapeutic treatment of psoriasis. Then, a deep neural network (DNN)-based drug-target interaction (DTI) model was trained by the DTI dataset to predict candidate molecular drugs. By considering adequate regulatory ability, toxicity, and sensitivity as drug design specifications, Naringin, Butein, and Betulinic acid were selected from the candidate molecular drugs and combined into potential multi-molecule drugs for the treatment of psoriasis.

Keywords: DNN-based DTI model; DTI databases; bid data mining; core signaling pathways; drug design specifications; multi-molecule drug; pathogenic mechanism of psoriasis; systems biology method.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Figure 1
Figure 1
Flowchart of systems biology methods to identify significant biomarkers of pathogenic mechanism as drug targets of psoriasis and the outline of systematic drug discovery design of psoriasis. Candidate GWGEN is composed of a protein–protein interaction network (PPIN) and gene regulatory network (GRN), which are constructed by PPIN and GRN datasets. Real GWGENs were obtained by pruning false positives from candidate GWGENs by microarray data GSE117468 through system order detection and system identification. The core GWGENs were extracted from real GWGENs by the PNP method. Core GWGENs of psoriasis and non-psoriasis are annotated by KEGG pathways to obtain core signaling pathways of psoriasis and non-psoriasis, respectively. The pathogenic biomarkers were identified for the pathogenesis of psoriasis by comparing the core signaling pathways of psoriasis and non-psoriasis. Finally, the potential drugs were discovered by the prediction of the DNN-based DTI model and the screening of drug design specifications. Then, the screened molecular drugs were combined as a multi-molecule drug for the therapeutic treatment of psoriasis.
Figure 2
Figure 2
(a) The real GWGEN of psoriasis; (b) the real GWGEN of non-psoriasis. The green lines represent the PPI, and the red lines represent the gene regulations. Numbers indicate the number of nodes.
Figure 3
Figure 3
(a) The core GWGEN of psoriasis; (b) the core GWGEN of non-psoriasis. The green lines represent the PPI, and the red lines represent the gene regulations. Numbers indicate the number of nodes.
Figure 4
Figure 4
The common and specific core signaling pathways and their downstream cellular dysfunctions between psoriasis and non-psoriasis. The figure shows the core genetic and epigenetic signaling pathways and pathogenic mechanisms of psoriasis. The left block contains the specific core signaling pathways of psoriasis. The middle block contains the overlapping core signaling pathways between psoriasis and non-psoriasis. The right block contains the specific core signaling pathways of non-psoriasis. The gene symbols in red or green font denote the selected significant biomarkers of the pathogenesis of psoriasis as drug targets of psoriasis.
Figure 5
Figure 5
The flowchart of design and discovery of a multi-molecule drug for therapeutic treatment of psoriasis. Drug-target interaction data were obtained from drug-target interaction databases. Then, the drug and target feature vectors were pre-processed, including downsampling, standardization, and PCA, respectively. After data preprocessing, drug target feature vectors were divided into training data and testing data for training the DNN-based DTI model. The well-trained DNN-based DTI model was used to predict candidate drugs for these biomarkers (drug targets). The potential molecule drugs were selected from predicted candidate drugs according to the drug design specifications and combined as multi-molecule drugs for the therapeutic treatment of psoriasis.
Figure 6
Figure 6
The training and validation accuracy (five-fold cross-validation). ”-o-” line in different colors denotes the training accuracy, “-◊-” line in different colors denotes the validation accuracy.
Figure 7
Figure 7
The training and validation loss (five-fold cross-validation).”-o-” line in different colors denotes the training loss, “-◊-” line in different colors denotes the validation loss.
Figure 8
Figure 8
The prediction performance of the trained DNN-based DTI model has an AUC score of 0.982 for its receiver operating characteristic (ROC) curve.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Lowes M.A., Bowcock A.M., Krueger J.G.J. Pathogenesis and therapy of psoriasis. Nature. 2007;445:866–873. doi: 10.1038/nature05663. - DOI - PubMed
    1. Magin P., Adams J., Heading G., Pond D., Smith W.J. The psychological sequelae of psoriasis: Results of a qualitative study. Psychol. Health Med. 2009;14:150–161. doi: 10.1080/13548500802512294. - DOI - PubMed
    1. Michalek I., Loring B., John S.J. Venereology. A systematic review of worldwide epidemiology of psoriasis. J. Eur. Acad. Dermatol. Venereol. 2017;31:205–212. doi: 10.1111/jdv.13854. - DOI - PubMed
    1. World Health Organization . World Health Organization Global Report on Psoriasis. World Health Organization; Geneva, Switzerland: 2016.
    1. Bhalerao J., Bowcock A.M. The genetics of psoriasis: A complex disorder of the skin and immune system. Hum. Mol. Genet. 1998;7:1537–1545. doi: 10.1093/hmg/7.10.1537. - DOI - PubMed

Grants and funding

This research received no external funding.