Dataset Information

Applications of Bayesian network models in predicting types of hematological malignancies.

ABSTRACT: Network analysis is the preferred approach for the detection of subtle but coordinated changes in expression of an interacting and related set of genes. We introduce a novel method based on the analyses of coexpression networks and Bayesian networks, and we use this new method to classify two types of hematological malignancies; namely, acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS). Our classifier has an accuracy of 93%, a precision of 98%, and a recall of 90% on the training dataset (n = 366); which outperforms the results reported by other scholars on the same dataset. Although our training dataset consists of microarray data, our model has a remarkable performance on the RNA-Seq test dataset (n = 74, accuracy = 89%, precision = 88%, recall = 98%), which confirms that eigengenes are robust with respect to expression profiling technology. These signatures are useful in classification and correctly predicting the diagnosis. They might also provide valuable information about the underlying biology of diseases. Our network analysis approach is generalizable and can be useful for classifying other diseases based on gene expression profiles. Our previously published Pigengene package is publicly available through Bioconductor, which can be used to conveniently fit a Bayesian network to gene expression data.

SUBMITTER: Agrahari R

PROVIDER: S-EPMC5934387 | biostudies-literature | 2018 May

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Applications of Bayesian network models in predicting types of hematological malignancies.

Agrahari Rupesh R Foroushani Amir A Docking T Roderick TR Chang Linda L Duns Gerben G Hudoba Monika M Karsan Aly A Zare Habil H

Scientific reports 20180503 1

Network analysis is the preferred approach for the detection of subtle but coordinated changes in expression of an interacting and related set of genes. We introduce a novel method based on the analyses of coexpression networks and Bayesian networks, and we use this new method to classify two types of hematological malignancies; namely, acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS). Our classifier has an accuracy of 93%, a precision of 98%, and a recall of 90% on the training d ...[more]

PMID: 29725024

Similar Datasets

Project description:Cellular immunotherapy has proven to be effective in the treatment of hematological cancers by donor lymphocyte infusion after allogeneic hematopoietic stem cell transplantation and more recently by targeted therapy with chimeric antigen or T-cell receptor-engineered T-cells. However, dependent on the tissue distribution of the antigens that are targeted, anti-tumor responses can be accompanied by undesired side effects. Therefore, detailed tissue distribution analysis is essential to estimate efficacy and toxicity of candidate targets for immunotherapy of hematological malignancies. In this study, we performed microarray gene expression analysis of hematological malignancies of different origins, healthy hematopoietic cells and various non-hematopoietic cell types from organs that are often targeted in detrimental immune responses after allogeneic stem cell transplantation leading to graft-versus-host disease. Non-hematopoietic cells were also cultured in the presence of IFN-Î³ to analyze gene expression under inflammatory circumstances. Gene expression was investigated by Illumina HT12.0 microarrays and quality control analysis was performed to confirm the cell-type origin and exclude contamination of non-hematopoietic cell samples with peripheral blood cells. Microarray data were validated by quantitative RT-PCR showing strong correlation between both platforms. Detailed gene expression profiles were generated for various minor histocompatibility antigens and B-cell surface antigens to illustrate the value of the microarray dataset to estimate efficacy and toxicity of candidate targets for immunotherapy. In conclusion, our microarray database provides a relevant platform to analyze and select candidate antigens with hematopoietic (lineage)-restricted expression as potential targets for immunotherapy of hematological cancers. Microarray gene expression analysis was performed on hematological malignancies of different origins and their healthy hematopoietic counterparts as well as on healthy non-hematopoietic cell types from organs that are often targeted in Graft-versus-Host Disease (166 samples in total). Various non-hematopoietic cell types were cultured in the presence of IFN-Î³ or T-cell culture supernatant to allow gene expression analysis under inflammatory conditions. Malignant and healthy hematopoietic cells were isolated by flow cytometry based on expression of specific surface markers and healthy non-hematopoietic cell types were isolated or cultured from tissue biopsies or surgically removed specimen. Various non-hematopoietic cell types were in vitro cultured for 4 days in the presence of IFN-Î³ or T-cell culture supernatant to mimic inflammation. The dataset allows comparison of gene expression between hematological malignancies and healthy non-hematopoietic cell types to estimate efficacy and toxicity of immunotherapeutic targets for hematological malignancies. Total RNA was isolated from (malignant) hematopoietic cells isolated by flow cytometry based on expression of specific surface markers and from healthy non-hematopoietic cell types isolated or cultured from tissue biopsies or surgically removed specimen. From various non-hematopoietic cell types, total RNA was also isolated after 4 days of in vitro culture in the presence of IFN-Î³ or T-cell culture supernatant.

Project description:Cellular immunotherapy has proven to be effective in the treatment of hematological cancers by donor lymphocyte infusion after allogeneic hematopoietic stem cell transplantation and more recently by targeted therapy with chimeric antigen or T-cell receptor-engineered T-cells. However, dependent on the tissue distribution of the antigens that are targeted, anti-tumor responses can be accompanied by undesired side effects. Therefore, detailed tissue distribution analysis is essential to estimate efficacy and toxicity of candidate targets for immunotherapy of hematological malignancies. In this study, we performed microarray gene expression analysis of hematological malignancies of different origins, healthy hematopoietic cells and various non-hematopoietic cell types from organs that are often targeted in detrimental immune responses after allogeneic stem cell transplantation leading to graft-versus-host disease. Non-hematopoietic cells were also cultured in the presence of IFN-γ to analyze gene expression under inflammatory circumstances. Gene expression was investigated by Illumina HT12.0 microarrays and quality control analysis was performed to confirm the cell-type origin and exclude contamination of non-hematopoietic cell samples with peripheral blood cells. Microarray data were validated by quantitative RT-PCR showing strong correlation between both platforms. Detailed gene expression profiles were generated for various minor histocompatibility antigens and B-cell surface antigens to illustrate the value of the microarray dataset to estimate efficacy and toxicity of candidate targets for immunotherapy. In conclusion, our microarray database provides a relevant platform to analyze and select candidate antigens with hematopoietic (lineage)-restricted expression as potential targets for immunotherapy of hematological cancers. Microarray gene expression analysis was performed on hematological malignancies of different origins and their healthy hematopoietic counterparts as well as on healthy non-hematopoietic cell types from organs that are often targeted in Graft-versus-Host Disease (166 samples in total). Various non-hematopoietic cell types were cultured in the presence of IFN-γ or T-cell culture supernatant to allow gene expression analysis under inflammatory conditions. Malignant and healthy hematopoietic cells were isolated by flow cytometry based on expression of specific surface markers and healthy non-hematopoietic cell types were isolated or cultured from tissue biopsies or surgically removed specimen. Various non-hematopoietic cell types were in vitro cultured for 4 days in the presence of IFN-γ or T-cell culture supernatant to mimic inflammation. The dataset allows comparison of gene expression between hematological malignancies and healthy non-hematopoietic cell types to estimate efficacy and toxicity of immunotherapeutic targets for hematological malignancies.

Dataset Information

Applications of Bayesian network models in predicting types of hematological malignancies.

Publications

Applications of Bayesian network models in predicting types of hematological malignancies.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets