Application of Genetic Algorithm-Based Support Vector Machine in Identification of Gene Expression Signatures for Psoriasis Classification: A Hybrid Model.
Journal
BioMed research international
ISSN: 2314-6141
Titre abrégé: Biomed Res Int
Pays: United States
ID NLM: 101600173
Informations de publication
Date de publication:
2021
2021
Historique:
received:
05
03
2021
accepted:
23
08
2021
entrez:
20
9
2021
pubmed:
21
9
2021
medline:
6
10
2021
Statut:
epublish
Résumé
Psoriasis is a chronic autoimmune disease impairing significantly the quality of life of the patient. The diagnosis of the disease is done via a visual inspection of the lesional skin by dermatologists. Classification of psoriasis using gene expression is an important issue for the early and effective treatment of the disease. Therefore, gene expression data and selection of suitable gene signatures are effective sources of information. We aimed to develop a hybrid classifier for the diagnosis of psoriasis based on two machine learning models of the genetic algorithm and support vector machine (SVM). The method also conducts gene signature selection. A publically available gene expression dataset was used to test the model. A number of 181 probe sets were selected among the original 54,675 probes using the hybrid model with a prediction accuracy of 100% over the test set. A number of 10 hub genes were identified using the protein-protein interaction network. Nine out of 10 identified genes were found in significant modules. The results showed that the genetic algorithm improved the SVM classifier performance significantly implying the ability of the proposed model in terms of detecting relevant gene expression signatures as the best features.
Sections du résumé
BACKGROUND
BACKGROUND
Psoriasis is a chronic autoimmune disease impairing significantly the quality of life of the patient. The diagnosis of the disease is done via a visual inspection of the lesional skin by dermatologists. Classification of psoriasis using gene expression is an important issue for the early and effective treatment of the disease. Therefore, gene expression data and selection of suitable gene signatures are effective sources of information.
METHODS
METHODS
We aimed to develop a hybrid classifier for the diagnosis of psoriasis based on two machine learning models of the genetic algorithm and support vector machine (SVM). The method also conducts gene signature selection. A publically available gene expression dataset was used to test the model.
RESULTS
RESULTS
A number of 181 probe sets were selected among the original 54,675 probes using the hybrid model with a prediction accuracy of 100% over the test set. A number of 10 hub genes were identified using the protein-protein interaction network. Nine out of 10 identified genes were found in significant modules.
CONCLUSIONS
CONCLUSIONS
The results showed that the genetic algorithm improved the SVM classifier performance significantly implying the ability of the proposed model in terms of detecting relevant gene expression signatures as the best features.
Identifiants
pubmed: 34540995
doi: 10.1155/2021/5520710
pmc: PMC8443357
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
5520710Informations de copyright
Copyright © 2021 Leili Tapak et al.
Déclaration de conflit d'intérêts
The authors declare that there is no conflict of interests.
Références
Nat Genet. 2009 Feb;41(2):199-204
pubmed: 19169254
J Invest Dermatol. 2014 Dec;134(12):2990-2993
pubmed: 24999591
J Mol Med (Berl). 2005 Dec;83(12):964-75
pubmed: 16283139
Gastroenterology. 2014 Sep;147(3):595-598.e5
pubmed: 24941021
Cleve Clin J Med. 2012 Jun;79(6):413-23
pubmed: 22660873
J Eur Acad Dermatol Venereol. 2020 May;34(5):1010-1018
pubmed: 31715033
World Neurosurg. 2022 Feb;158:e543-e556
pubmed: 34775094
Oncotarget. 2016 Feb 2;7(5):5800-14
pubmed: 26735579
Proc Natl Acad Sci U S A. 2005 Feb 8;102(6):2075-80
pubmed: 15671179
Eur J Pharm Sci. 2020 Oct 1;153:105465
pubmed: 32668312
Nucleic Acids Res. 2019 Jan 8;47(D1):D607-D613
pubmed: 30476243
Genome Biol. 2015 Mar 25;16:58
pubmed: 25885816
Cytotechnology. 2018 Oct;70(5):1315-1323
pubmed: 29802490
Actas Dermosifiliogr (Engl Ed). 2019 Jan - Feb;110(1):12-19
pubmed: 30509759
Am J Hum Genet. 2019 Nov 7;105(5):1040-1047
pubmed: 31630789
J Invest Dermatol. 2009 Oct;129(10):2411-8
pubmed: 19458634
J Cell Mol Med. 2019 Mar;23(3):2207-2218
pubmed: 30661291
Arch Biochem Biophys. 2015 Oct 1;583:130-9
pubmed: 26297660
PLoS One. 2011;6(8):e22258
pubmed: 21857917
Ann Rheum Dis. 2005 Mar;64 Suppl 2:ii30-6
pubmed: 15708932
Clin Proteomics. 2020 Aug 12;17:30
pubmed: 32817748
PLoS One. 2013 Jul 16;8(7):e68667
pubmed: 23874713
Am J Cancer Res. 2020 Sep 01;10(9):2768-2784
pubmed: 33042616
Curr Opin Rheumatol. 2004 Jul;16(4):331-7
pubmed: 15201593
Plant Physiol. 2012 Apr;158(4):1819-32
pubmed: 22331412
Genome Res. 2003 Nov;13(11):2498-504
pubmed: 14597658
Hum Mol Genet. 2001 Aug 15;10(17):1793-805
pubmed: 11532989
Sci Rep. 2020 Jun 17;10(1):9868
pubmed: 32555320
PLoS One. 2011 Apr 04;6(4):e18266
pubmed: 21483750
Actas Dermosifiliogr. 2012 Dec;103(10):853-62
pubmed: 23157913
Oncol Lett. 2017 Jul;14(1):41-46
pubmed: 28693133
J Eur Acad Dermatol Venereol. 2017 Dec;31(12):1999-2009
pubmed: 28681405
Elife. 2019 Dec 16;8:
pubmed: 31841110
BMC Cancer. 2013 Jun 22;13:303
pubmed: 23800275
Genomics. 2014 Jan;103(1):48-55
pubmed: 24239985
G Ital Dermatol Venereol. 2012 Apr;147(2):189-202
pubmed: 22481582
J Invest Dermatol. 2009 Dec;129(12):2795-804
pubmed: 19571819
FEBS Lett. 2019 Aug;593(15):2060-2068
pubmed: 31180576
J Infect Dis. 2016 Dec 1;214(11):1762-1772
pubmed: 27658692
J Invest Dermatol. 2007 Jan;127(1):163-9
pubmed: 16858420
Ann Rheum Dis. 2005 Mar;64 Suppl 2:ii18-23; discussion ii24-5
pubmed: 15708928
Int J Mol Sci. 2019 Mar 23;20(6):
pubmed: 30909615
Scand J Immunol. 2015 Feb;81(2):96-101
pubmed: 25410188
Biomed Res Int. 2015;2015:292683
pubmed: 26504797