College of Science and Engineering
Permanent URI for this collectionhttps://hdl.handle.net/10877/17053
Browse
Browsing College of Science and Engineering by Author "Agrahari, Rupesh"
Now showing 1 - 2 of 2
Results Per Page
Sort Options
Item Applications of Bayesian Network Models in Predicting Types of Hematological Malignancies(Nature Publishing Group, 2018-05) Agrahari, Rupesh; Foroushani, Amir; Docking, T. Roderick; Chang, Linda; Duns, Gerben; Hudoba, Monika; Karsan, Aly; Zare, HabilNetwork 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.Item Large-Scale Gene Network Analysis Reveals the Significance of Extracellular Matrix Pathway and Homeobox Genes in Acute Myeloid Leukemia: An Introduction to the Pigengene Package and its Applications(BioMed Central, 2017-01) Foroushani, Amir; Agrahari, Rupesh; Docking, T. Roderick; Chang, Linda; Duns, Gerben; Hudoba, Monika; Karsan, Aly; Zare, HabilBackground: The distinct types of hematological malignancies have different biological mechanisms and prognoses. For instance, myelodysplastic syndrome (MDS) is generally indolent and low risk; however, it may transform into acute myeloid leukemia (AML), which is much more aggressive. Methods: We develop a novel network analysis approach that uses expression of eigengenes to delineate the biological differences between these two diseases. Results: We find that specific genes in the extracellular matrix pathway are underexpressed in AML. We validate this finding in three ways: (a) We train our model on a microarray dataset of 364 cases and test it on an RNA Seq dataset of 74 cases. Our model showed 95% sensitivity and 86% specificity in the training dataset and showed 98% sensitivity and 91% specificity in the test dataset. This confirms that the identified biological signatures are independent from the expression profiling technology and independent from the training dataset. (b) Immunocytochemistry confirms that MMP9, an exemplar protein in the extracellular matrix, is underexpressed in AML. (c) MMP9 is hypermethylated in the majority of AML cases (n=194, Welch's t-test p-value <10-138), which complies with its low expression in AML. Our novel network analysis approach is generalizable and useful in studying other complex diseases (e.g., breast cancer prognosis). We implement our methodology in the Pigengene software package, which is publicly available through Bioconductor. Conclusions: Eigengenes define informative biological signatures that are robust with respect to expression profiling technology. These signatures provide valuable information about the underlying biology of diseases, and they are useful in predicting diagnosis and prognosis.