Browsing by Author "Yazici, Miray Unlu"

Now showing 1 - 5 of 5

In-Silico Methods to Identify Common MicroRNAs and Pathways of Neuromuscular Diseases
(IEEE, 345 E 47TH ST, NEW YORK, NY 10017 USA, 2019) Yazici, Miray Unlu; Menges, Evrim Aksu; Ulum, Yeliz Z. Akkaya; Hayta, Burcu Balci; Bakir-Gungor, Burcu; AGÜ, Yaşam ve Doğa Bilimleri Fakültesi, Biyomühendislik Bölümü
Neuromuscular disorders (NMD) are a heterogeneous group of diseases characterized by the loss of function of the peripheral nerves and muscles. However, there are no effective and widespread therapeutic approaches to prevent or delay the progression of these disease types. microRNAs (miRNAs) which cause significant changes in gene expression by binding to target messenger RNAs (mRNAs), are known to have an effect on disease mechanisms. In this study, by integrating different bioinformatics methods, we aim to find miRNAs, target genes and pathways related to a group of neuromuscular diseases. For this purpose, we determined 17 miRNAs that show significant expression changes between patient and healthy groups; predicted target genes of these miRNAs; and identified affected pathways using subnetwork discovery, functional enrichment based algorithms. In our study, we integrated different in-silico approaches that proceed in top-down manner or bottom-up manner. The identified candidate miRNAs, genes and pathways, which could help to explain neuromuscular disease development mechanisms, are now under investigation in wet-lab.
Integrative analyses in omics data: Machine learning perspective
(Deutsche Gesellschaft fur Medizinische Informatik, Biometrie und Epidemiologie e.V., 2023) Yazici, Miray Unlu; Bakir-Gungor, Burcu; Yousef, Malik; 0000-0001-8165-6164; 0000-0002-2272-6270; AGÜ, Yaşam ve Doğa Bilimleri Fakültesi, Biyomühendislik Bölümü; Yazici, Miray Unlu; Bakir-Gungor, Burcu
Developments in the high throughput technologies have enabled the production of an immense amount of knowledge at the multi-omics level. Considering complex diseases which are affected by multi-factors, single omics datasets might not be sufficient to unveil the molecular mechanisms of heterogeneous diseases. Providing a comprehensive and systematic overview to explain disease hallmarks in significant depth is critical. Utilizing multi-omics datasets has led to the development of a variety of tools and platforms. Machine learning models are utilized in a wide variety of tools to tackle the complexity of disorders and to identify new biomolecular signatures and potential markers. Underlying aspects of these approaches are based on training the models for making predictions and classification of the given data. In this review, we describe current machine learning-based approaches and available implementations. Challenges in the enlightenment of disease mechanisms of onset and progression and future development of the field of medicine will be discussed. The prominence of biological interpretation of model output with corresponding biological knowledge will be also covered in this review.
Invention of 3Mint for feature grouping and scoring in multi-omics
(FRONTIERS MEDIA SA, 2023) Yazici, Miray Unlu; Marron, J. S.; Bakir-Gungor, Burcu; Zou, Fei; Yousef, Malik; 0000-0002-2272-6270; AGÜ, Mühendislik Fakültesi, Bilgisayar Mühendisliği Bölümü; Bakir-Gungor, Burcu
Advanced genomic and molecular profiling technologies accelerated the enlightenment of the regulatory mechanisms behind cancer development and progression, and the targeted therapies in patients. Along this line, intense studies with immense amounts of biological information have boosted the discovery of molecular biomarkers. Cancer is one of the leading causes of death around the world in recent years. Elucidation of genomic and epigenetic factors in Breast Cancer (BRCA) can provide a roadmap to uncover the disease mechanisms. Accordingly, unraveling the possible systematic connections between-omics data types and their contribution to BRCA tumor progression is crucial. In this study, we have developed a novel machine learning (ML) based integrative approach for multi-omics data analysis. This integrative approach combines information from gene expression (mRNA), microRNA (miRNA) and methylation data. Due to the complexity of cancer, this integrated data is expected to improve the prediction, diagnosis and treatment of disease through patterns only available from the 3-way interactions between these 3-omics datasets. In addition, the proposed method bridges the interpretation gap between the disease mechanisms that drive onset and progression. Our fundamental contribution is the 3 Multi-omics integrative tool (3Mint). This tool aims to perform grouping and scoring of groups using biological knowledge. Another major goal is improved gene selection via detection of novel groups of cross-omics biomarkers. Performance of 3Mint is assessed using different metrics. Our computational performance evaluations showed that the 3Mint classifies the BRCA molecular subtypes with lower number of genes when compared to the miRcorrNet tool which uses miRNA and mRNA gene expression profiles in terms of similar performance metrics (95% Accuracy). The incorporation of methylation data in 3Mint yields a much more focused analysis. The 3Mint tool and all other supplementary files are available at .
miRcorrNetPro: Unraveling Algorithmic Insights through Cross-Validation in Multi-Omics Integration for Comprehensive Data Analysis
(Institute of Electrical and Electronics Engineers Inc., 2023) Yazici, Miray Unlu; Yousef, Malik; Marron J.S.; Bakir-Gungor, Burcu; 0000-0001-8165-6164; 0000-0002-2272-6270; AGÜ, Yaşam ve Doğa Bilimleri Fakültesi, Biyomühendislik Bölümü; Yazici, Miray Unlu; Bakir-Gungor, Burcu
High throughput -omics technologies facilitate the investigation of regulatory mechanisms of complex diseases. Along this line, scientists develop promising tools and methods to extend our understanding at the molecular and functional levels. To this end, miRcorrNet tool performs integrative analysis of microRNA (miRNA) and gene expression profiles via machine learning (ML) approach to identify significant miRNA groups and their associated target genes. In this study, we propose miRcorrNetPro tool, which extends miRcorrNet by tracking group scoring, ranking and other information through the cross-validation iterations. Heatmap visualizations enable deep novel insights into the collective behavior of clusters of groups in cellular signaling and hence facilitate detection of potential biomarkers for the disease under investigation. Although miRcorrNetPro is designed as a generic tool, here we present our findings and potential miRNA biomarkers for Breast Cancer (BRCA). The miRcorrNetPro tool and all other supplementary files are available at https://github.com/MirayUnlu/miRcorrNetPro.
A New Method to Identify Affected Pathway Subnetworks and Clusters in Colon Cancer
(IEEE, 345 E 47TH ST, NEW YORK, NY 10017 USA, 11.09.2019) Goy, Gokhan; Yazici, Miray Unlu; Bakir-Gungor, Buren; AGÜ, Mühendislik Fakültesi, Bilgisayar Mühendisliği Bölümü
Nowadays new technological developments that play an important role in the production of big data have brought about the interpretation, sharing and storage of data related to complex diseases. Combining multi-omic data in different molecular levels is potentially important for understanding the biological origin of complex diseases. One of these complex diseases is cancer of different types, which has one of the highest causes of death worldwide. The integration of multiple omic data in the framework of a comprehensive analysis and identification of relevant pathways contribute to the development of therapeutic approaches related to disease. In this study, RNA and methylation data (genes and p values) of colon adenocarcinoma were obtained from TCGA data portal and combined with Fisher's method. While protein subnetworks affected by the disease were identified by using subnetwork algorithm, pathways related to the disease and genes associated with these pathways were determined by functional enrichment analysis. Using gene-pathway relationship matrix, kappa scores of pathways were determined by similarity calculation. In this way, the pathways were clustered according to the hierarchically optimal number, as a result, the most important pathway clusters and related genes that are effective in disease formation identified.