oru.sePublikationer
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Studying Microarray Gene Expression Data of Schizophrenic Patients for Derivation of a Diagnostic Signature through the Aid of Machine Learning
Örebro University, School of Medical Sciences. Metabolic Engineering and Bioinformatics Group, Institute of Biology, Medicinal Chemistry and Biotechnology, National Hellenic Research Foundation, Athens, Greece.
Metabolic Engineering and Bioinformatics Group, Institute of Biology, Medicinal Chemistry and Biotechnology, National Hellenic Research Foundation, Athens, Greece; e-NIOS Applications PC, Athens, Greece .
Örebro University, School of Health Sciences.ORCID iD: 0000-0001-8102-1804
Laboratory of Biotechnology, School of Chemical Engineering, National Technical University of Athens, Athens, Greece.
Show others and affiliations
2016 (English)In: Biometrics & Biostatistics International Journal, ISSN 2378-315X, Vol. 4, no 5, 00106Article in journal (Refereed) Published
Abstract [en]

Schizophrenia is a complex psychiatric disease that is affected by multiple genes, some of which could be used as biomarkers for specific diagnosis of the disease. In this work, we explore the power of machine learning methodologies for predicting schizophrenia, through the derivation of a biomarker gene signature for robust diagnostic classification purposes. Postmortem brain gene expression data from the anterior prefrontal cortex of schizophrenia patients were used as training data for the construction of the classifiers. Several machine learning algorithms, such as support vector machines, random forests, and extremely randomized trees classifiers were developed and their performance was tested. After applying the feature selection method of support vector machines recursive feature elimination a 21-gene model was derived. Using these genes for developing classification models, the random forests algorithm outperformed all examined algorithms achieving an area under the curve of 0.98 and sensitivity of 0.89, discriminating schizophrenia from healthy control samples with high efficiency. The 21-gene model that was derived from the feature selection is suggested for classifying schizophrenic patients, as it was successfully applied on an independent dataset of postmortem brain samples from the superior temporal cortex, and resulted in a classification model that achieved an area under the curve score of 0.91. Additionally, the functional analysis of the statistically significant genes indicated many mechanisms related to the immune system.

Place, publisher, year, edition, pages
MedCrave , 2016. Vol. 4, no 5, 00106
Keyword [en]
Classification, Schizophrenia, Machine learning, Gene expression, Microarray studies, Support vector machines, Adaboost
National Category
Other Basic Medicine
Research subject
Biomedicine
Identifiers
URN: urn:nbn:se:oru:diva-53536OAI: oai:DiVA.org:oru-53536DiVA: diva2:1047259
Note

DOI 10.15406/bbij.2016.04.00106

Available from: 2016-11-17 Created: 2016-11-17 Last updated: 2017-10-18Bibliographically approved
In thesis
1. Integration of functional genomics and data mining methodologies in the study of bipolar disorder and schizophrenia
Open this publication in new window or tab >>Integration of functional genomics and data mining methodologies in the study of bipolar disorder and schizophrenia
2016 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

Bipolar disorder and schizophrenia are two severe psychiatric disorders characterized by a complex genetic basis, coupled to the influence of environmental factors. In this thesis, functional genomic analysis tools were used for the study of the underlying pathophysiology of these disorders, focusing on gene expression and function on a global scale with the application of high-throughput methods. Datasets from public databases regarding transcriptomic data of postmortem brain and skin fibroblast cells of patients with either schizophrenia or bipolar disorder were analyzed in order to identify differentially expressed genes. In addition, fibroblast cells of bipolar disorder patients obtained from the Biobank of the Neuropsychiatric Research Laboratory of Örebro University were cultured, RNA was extracted and used for microarray analysis. In order to gain deeper insight into the biological mechanisms related to the studied psychiatric disorders, the differentially expressed gene lists were subjected to pathway and target prioritization analysis, using proprietary tools developed by the group of Metabolic Engineering and Bioinformatics, of the National Hellenic Research Foundation, thus indicating various cellular processes as significantly altered. Many of the molecular processes derived from the analysis of the postmortem brain data of schizophrenia and bipolar disorder were also identified in the skin fibroblast cells. Additionally, through the use of machine learning methods, gene expression data from patients with schizophrenia were exploited for the identification of a subset of genes with discriminative ability between schizophrenia and healthy control subjects. Interestingly, a set of genes with high separating efficiency was derived from fibroblast gene expression profiling. This thesis suggests the suitability of skin fibroblasts as a reliable model for the diagnostic evaluation of psychiatric disorders and schizophrenia in particular, through the construction of promising machine-learning based classification models, exploiting gene expression data from peripheral tissues.

Place, publisher, year, edition, pages
Örebro: Örebro university, 2016. 98 p.
Series
Örebro Studies in Medicine, ISSN 1652-4063 ; 153
Keyword
Bipolar Disorder, Schizophrenia, Fibroblasts, DNA Microarrays, Machine Learning, Functional Analysis, Gene Expression, Transcriptomics
National Category
Other Basic Medicine
Identifiers
urn:nbn:se:oru:diva-52644 (URN)978-91-7529-168-0 (ISBN)
Public defence
2016-12-09, Campus USÖ, hörsal C3, Södra Grev Rosengatan 32, Örebro, 09:00 (English)
Opponent
Supervisors
Available from: 2016-09-28 Created: 2016-09-28 Last updated: 2017-10-17Bibliographically approved

Open Access in DiVA

No full text

Other links

Studying Microarray Gene Expression Data of Schizophrenic Patients for Derivation of a Diagnostic Signature through the Aid of Machine Learning

Search in DiVA

By author/editor
Logotheti, MarianthiVenizelos, Nikolaos
By organisation
School of Medical SciencesSchool of Health Sciences
Other Basic Medicine

Search outside of DiVA

GoogleGoogle Scholar

Total: 320 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf