Diagnostic classification of schizophrenia by neural network analysis of blood-based gene expression signatures

Makoto Takahashi, Hiroshi Hayashi, Yuichiro Watanabe, Kazushi Sawamura, Naoki Fukui, Junzo Watanabe, Tsuyoshi Kitajima, Yoshio Yamanouchi, Nakao Iwata, Katsuyoshi Mizukami, Takafumi Hori, Kazutaka Shimoda, Hiroshi Ujike, Norio Ozaki, Kentarou Iijima, Kazuo Takemura, Hideyuki Aoshima, Toshiyuki Someya

Research output: Contribution to journalArticle

49 Citations (Scopus)

Abstract

Gene expression profiling with microarray technology suggests that peripheral blood cells might be a surrogate for postmortem brain tissue in studies of schizophrenia. The development of an accessible peripheral biomarker would substantially help in the diagnosis of this disease. We used a bioinformatics approach to examine whether the gene expression signature in whole blood contains enough information to make a specific diagnosis of schizophrenia. Unpaired t-tests of gene expression datasets from 52 antipsychotics-free schizophrenia patients and 49 normal controls identified 792 differentially expressed probes. Functional profiling with DAVID revealed that eleven of these genes were previously reported to be associated with schizophrenia, and 73 of them were expressed in the brain tissue. We analyzed the datasets with one of the supervised classifiers, artificial neural networks (ANNs). The samples were subdivided into training and testing sets. Quality filtering and stepwise forward selection identified 14 probes as predictors of the diagnosis. ANNs were then trained with the selected probes as the input and the training set for known diagnosis as the output. The constructed model achieved 91.2% diagnostic accuracy in the training set and 87.9% accuracy in the hold-out testing set. On the other hand, hierarchical clustering, a standard but unsupervised classifier, failed to separate patients and controls. These results suggest analysis of a blood-based gene expression signature with the supervised classifier, ANNs, might be a diagnostic tool for schizophrenia.

Original languageEnglish
Pages (from-to)210-218
Number of pages9
JournalSchizophrenia Research
Volume119
Issue number1-3
DOIs
Publication statusPublished - 01-06-2010

Fingerprint

Transcriptome
Schizophrenia
Brain
Gene Expression Profiling
Computational Biology
Antipsychotic Agents
Cluster Analysis
Blood Cells
Biomarkers
Technology
Gene Expression
Genes
Datasets

All Science Journal Classification (ASJC) codes

  • Psychiatry and Mental health
  • Biological Psychiatry

Cite this

Takahashi, M., Hayashi, H., Watanabe, Y., Sawamura, K., Fukui, N., Watanabe, J., ... Someya, T. (2010). Diagnostic classification of schizophrenia by neural network analysis of blood-based gene expression signatures. Schizophrenia Research, 119(1-3), 210-218. https://doi.org/10.1016/j.schres.2009.12.024
Takahashi, Makoto ; Hayashi, Hiroshi ; Watanabe, Yuichiro ; Sawamura, Kazushi ; Fukui, Naoki ; Watanabe, Junzo ; Kitajima, Tsuyoshi ; Yamanouchi, Yoshio ; Iwata, Nakao ; Mizukami, Katsuyoshi ; Hori, Takafumi ; Shimoda, Kazutaka ; Ujike, Hiroshi ; Ozaki, Norio ; Iijima, Kentarou ; Takemura, Kazuo ; Aoshima, Hideyuki ; Someya, Toshiyuki. / Diagnostic classification of schizophrenia by neural network analysis of blood-based gene expression signatures. In: Schizophrenia Research. 2010 ; Vol. 119, No. 1-3. pp. 210-218.
@article{1fa4c9ecc95d46d9977576929573c260,
title = "Diagnostic classification of schizophrenia by neural network analysis of blood-based gene expression signatures",
abstract = "Gene expression profiling with microarray technology suggests that peripheral blood cells might be a surrogate for postmortem brain tissue in studies of schizophrenia. The development of an accessible peripheral biomarker would substantially help in the diagnosis of this disease. We used a bioinformatics approach to examine whether the gene expression signature in whole blood contains enough information to make a specific diagnosis of schizophrenia. Unpaired t-tests of gene expression datasets from 52 antipsychotics-free schizophrenia patients and 49 normal controls identified 792 differentially expressed probes. Functional profiling with DAVID revealed that eleven of these genes were previously reported to be associated with schizophrenia, and 73 of them were expressed in the brain tissue. We analyzed the datasets with one of the supervised classifiers, artificial neural networks (ANNs). The samples were subdivided into training and testing sets. Quality filtering and stepwise forward selection identified 14 probes as predictors of the diagnosis. ANNs were then trained with the selected probes as the input and the training set for known diagnosis as the output. The constructed model achieved 91.2{\%} diagnostic accuracy in the training set and 87.9{\%} accuracy in the hold-out testing set. On the other hand, hierarchical clustering, a standard but unsupervised classifier, failed to separate patients and controls. These results suggest analysis of a blood-based gene expression signature with the supervised classifier, ANNs, might be a diagnostic tool for schizophrenia.",
author = "Makoto Takahashi and Hiroshi Hayashi and Yuichiro Watanabe and Kazushi Sawamura and Naoki Fukui and Junzo Watanabe and Tsuyoshi Kitajima and Yoshio Yamanouchi and Nakao Iwata and Katsuyoshi Mizukami and Takafumi Hori and Kazutaka Shimoda and Hiroshi Ujike and Norio Ozaki and Kentarou Iijima and Kazuo Takemura and Hideyuki Aoshima and Toshiyuki Someya",
year = "2010",
month = "6",
day = "1",
doi = "10.1016/j.schres.2009.12.024",
language = "English",
volume = "119",
pages = "210--218",
journal = "Schizophrenia Research",
issn = "0920-9964",
publisher = "Elsevier",
number = "1-3",

}

Takahashi, M, Hayashi, H, Watanabe, Y, Sawamura, K, Fukui, N, Watanabe, J, Kitajima, T, Yamanouchi, Y, Iwata, N, Mizukami, K, Hori, T, Shimoda, K, Ujike, H, Ozaki, N, Iijima, K, Takemura, K, Aoshima, H & Someya, T 2010, 'Diagnostic classification of schizophrenia by neural network analysis of blood-based gene expression signatures', Schizophrenia Research, vol. 119, no. 1-3, pp. 210-218. https://doi.org/10.1016/j.schres.2009.12.024

Diagnostic classification of schizophrenia by neural network analysis of blood-based gene expression signatures. / Takahashi, Makoto; Hayashi, Hiroshi; Watanabe, Yuichiro; Sawamura, Kazushi; Fukui, Naoki; Watanabe, Junzo; Kitajima, Tsuyoshi; Yamanouchi, Yoshio; Iwata, Nakao; Mizukami, Katsuyoshi; Hori, Takafumi; Shimoda, Kazutaka; Ujike, Hiroshi; Ozaki, Norio; Iijima, Kentarou; Takemura, Kazuo; Aoshima, Hideyuki; Someya, Toshiyuki.

In: Schizophrenia Research, Vol. 119, No. 1-3, 01.06.2010, p. 210-218.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Diagnostic classification of schizophrenia by neural network analysis of blood-based gene expression signatures

AU - Takahashi, Makoto

AU - Hayashi, Hiroshi

AU - Watanabe, Yuichiro

AU - Sawamura, Kazushi

AU - Fukui, Naoki

AU - Watanabe, Junzo

AU - Kitajima, Tsuyoshi

AU - Yamanouchi, Yoshio

AU - Iwata, Nakao

AU - Mizukami, Katsuyoshi

AU - Hori, Takafumi

AU - Shimoda, Kazutaka

AU - Ujike, Hiroshi

AU - Ozaki, Norio

AU - Iijima, Kentarou

AU - Takemura, Kazuo

AU - Aoshima, Hideyuki

AU - Someya, Toshiyuki

PY - 2010/6/1

Y1 - 2010/6/1

N2 - Gene expression profiling with microarray technology suggests that peripheral blood cells might be a surrogate for postmortem brain tissue in studies of schizophrenia. The development of an accessible peripheral biomarker would substantially help in the diagnosis of this disease. We used a bioinformatics approach to examine whether the gene expression signature in whole blood contains enough information to make a specific diagnosis of schizophrenia. Unpaired t-tests of gene expression datasets from 52 antipsychotics-free schizophrenia patients and 49 normal controls identified 792 differentially expressed probes. Functional profiling with DAVID revealed that eleven of these genes were previously reported to be associated with schizophrenia, and 73 of them were expressed in the brain tissue. We analyzed the datasets with one of the supervised classifiers, artificial neural networks (ANNs). The samples were subdivided into training and testing sets. Quality filtering and stepwise forward selection identified 14 probes as predictors of the diagnosis. ANNs were then trained with the selected probes as the input and the training set for known diagnosis as the output. The constructed model achieved 91.2% diagnostic accuracy in the training set and 87.9% accuracy in the hold-out testing set. On the other hand, hierarchical clustering, a standard but unsupervised classifier, failed to separate patients and controls. These results suggest analysis of a blood-based gene expression signature with the supervised classifier, ANNs, might be a diagnostic tool for schizophrenia.

AB - Gene expression profiling with microarray technology suggests that peripheral blood cells might be a surrogate for postmortem brain tissue in studies of schizophrenia. The development of an accessible peripheral biomarker would substantially help in the diagnosis of this disease. We used a bioinformatics approach to examine whether the gene expression signature in whole blood contains enough information to make a specific diagnosis of schizophrenia. Unpaired t-tests of gene expression datasets from 52 antipsychotics-free schizophrenia patients and 49 normal controls identified 792 differentially expressed probes. Functional profiling with DAVID revealed that eleven of these genes were previously reported to be associated with schizophrenia, and 73 of them were expressed in the brain tissue. We analyzed the datasets with one of the supervised classifiers, artificial neural networks (ANNs). The samples were subdivided into training and testing sets. Quality filtering and stepwise forward selection identified 14 probes as predictors of the diagnosis. ANNs were then trained with the selected probes as the input and the training set for known diagnosis as the output. The constructed model achieved 91.2% diagnostic accuracy in the training set and 87.9% accuracy in the hold-out testing set. On the other hand, hierarchical clustering, a standard but unsupervised classifier, failed to separate patients and controls. These results suggest analysis of a blood-based gene expression signature with the supervised classifier, ANNs, might be a diagnostic tool for schizophrenia.

UR - http://www.scopus.com/inward/record.url?scp=77952674096&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77952674096&partnerID=8YFLogxK

U2 - 10.1016/j.schres.2009.12.024

DO - 10.1016/j.schres.2009.12.024

M3 - Article

VL - 119

SP - 210

EP - 218

JO - Schizophrenia Research

JF - Schizophrenia Research

SN - 0920-9964

IS - 1-3

ER -