Development and validation of a skin fibroblast biomarker profile for schizophrenic patients
Gene expression profiles of non-neural tissues through microarray technology could be used in schizophrenia studies, adding more information to the results from similar studies on postmortem brain tissue. The ultimate goal of such studies is to develop accessible biomarkers. Supervised machine learn...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
AIMS Press
2016-12-01
|
| Series: | AIMS Bioengineering |
| Subjects: | |
| Online Access: | http://www.aimspress.com/Bioengineering/article/1160/fulltext.html |
| _version_ | 1819154153504505856 |
|---|---|
| author | Marianthi Logotheti Eleftherios Pilalis Nikolaos Venizelos Fragiskos Kolisis Aristotelis Chatziioannou |
| author_facet | Marianthi Logotheti Eleftherios Pilalis Nikolaos Venizelos Fragiskos Kolisis Aristotelis Chatziioannou |
| author_sort | Marianthi Logotheti |
| collection | DOAJ |
| description | Gene expression profiles of non-neural tissues through microarray technology could be used in schizophrenia studies, adding more information to the results from similar studies on postmortem brain tissue. The ultimate goal of such studies is to develop accessible biomarkers. Supervised machine learning methodologies were used, in order to examine if the gene expression from skin fibroblast cells could be exploited for the classification of schizophrenic subjects. A dataset of skin fibroblasts gene expression of schizophrenia patients was obtained from Gene Expression Omnibus database. After applying statistical criteria, we concluded to genes that present a differential expression between the schizophrenic patients and the healthy controls. Based on those genes, functional profiling was performed with the BioInfoMiner web tool. After the statistical analysis, 63 genes were identified as differentially expressed. The functional profiling revealed interesting terms and pathways, such as mitogen activated protein kinase and cyclic adenosine monophosphate signaling pathways, as well as immune-related mechanisms. A subset of 16 differentially expressed genes from fibroblast gene expression profiling that occurred after Support Vector Machines Recursive Feature Elimination could efficiently separate schizophrenic from healthy controls subjects. These findings suggest that through the analysis of fibroblast based gene expression signature and with the application of machine learning methodologies we might conclude to a diagnostic classification model in schizophrenia. |
| first_indexed | 2024-12-22T15:16:33Z |
| format | Article |
| id | doaj.art-0c7ae86d69a34a5fb8049cc381e8e11f |
| institution | Directory Open Access Journal |
| issn | 2375-1495 |
| language | English |
| last_indexed | 2024-12-22T15:16:33Z |
| publishDate | 2016-12-01 |
| publisher | AIMS Press |
| record_format | Article |
| series | AIMS Bioengineering |
| spelling | doaj.art-0c7ae86d69a34a5fb8049cc381e8e11f2022-12-21T18:21:44ZengAIMS PressAIMS Bioengineering2375-14952016-12-013455256510.3934/bioeng.2016.4.552bioeng-03-00552Development and validation of a skin fibroblast biomarker profile for schizophrenic patientsMarianthi LogothetiEleftherios PilalisNikolaos Venizelos0Fragiskos Kolisis1Aristotelis ChatziioannouNeuropsychiatric Research Laboratory, Faculty of Medicine and Health, School of Health and Medical Sciences, Örebro University, Örebro, SwedenLaboratory of Biotechnology, School of Chemical Engineering, National Technical University of Athens, Athens, GreeceGene expression profiles of non-neural tissues through microarray technology could be used in schizophrenia studies, adding more information to the results from similar studies on postmortem brain tissue. The ultimate goal of such studies is to develop accessible biomarkers. Supervised machine learning methodologies were used, in order to examine if the gene expression from skin fibroblast cells could be exploited for the classification of schizophrenic subjects. A dataset of skin fibroblasts gene expression of schizophrenia patients was obtained from Gene Expression Omnibus database. After applying statistical criteria, we concluded to genes that present a differential expression between the schizophrenic patients and the healthy controls. Based on those genes, functional profiling was performed with the BioInfoMiner web tool. After the statistical analysis, 63 genes were identified as differentially expressed. The functional profiling revealed interesting terms and pathways, such as mitogen activated protein kinase and cyclic adenosine monophosphate signaling pathways, as well as immune-related mechanisms. A subset of 16 differentially expressed genes from fibroblast gene expression profiling that occurred after Support Vector Machines Recursive Feature Elimination could efficiently separate schizophrenic from healthy controls subjects. These findings suggest that through the analysis of fibroblast based gene expression signature and with the application of machine learning methodologies we might conclude to a diagnostic classification model in schizophrenia.http://www.aimspress.com/Bioengineering/article/1160/fulltext.htmlschizophreniaperipheral biomarkerfibroblastsmachine learningclassification |
| spellingShingle | Marianthi Logotheti Eleftherios Pilalis Nikolaos Venizelos Fragiskos Kolisis Aristotelis Chatziioannou Development and validation of a skin fibroblast biomarker profile for schizophrenic patients AIMS Bioengineering schizophrenia peripheral biomarker fibroblasts machine learning classification |
| title | Development and validation of a skin fibroblast biomarker profile for schizophrenic patients |
| title_full | Development and validation of a skin fibroblast biomarker profile for schizophrenic patients |
| title_fullStr | Development and validation of a skin fibroblast biomarker profile for schizophrenic patients |
| title_full_unstemmed | Development and validation of a skin fibroblast biomarker profile for schizophrenic patients |
| title_short | Development and validation of a skin fibroblast biomarker profile for schizophrenic patients |
| title_sort | development and validation of a skin fibroblast biomarker profile for schizophrenic patients |
| topic | schizophrenia peripheral biomarker fibroblasts machine learning classification |
| url | http://www.aimspress.com/Bioengineering/article/1160/fulltext.html |
| work_keys_str_mv | AT marianthilogotheti developmentandvalidationofaskinfibroblastbiomarkerprofileforschizophrenicpatients AT eleftheriospilalis developmentandvalidationofaskinfibroblastbiomarkerprofileforschizophrenicpatients AT nikolaosvenizelos developmentandvalidationofaskinfibroblastbiomarkerprofileforschizophrenicpatients AT fragiskoskolisis developmentandvalidationofaskinfibroblastbiomarkerprofileforschizophrenicpatients AT aristotelischatziioannou developmentandvalidationofaskinfibroblastbiomarkerprofileforschizophrenicpatients |