3Mont

Last edited  Mar 23, 2024 11:14 PM

ENTM-TS

Last edited  Dec 29, 2023 8:39 AM

GediNET

Last edited  Apr 25, 2022 10:28 PM

Malik Yousef and Burcu Bakir-Gungor Research Group

dwds

Last edited  Jun 25, 2020 11:52 PM

miRdisNET

Last edited  Feb 6, 2023 8:24 AM

miRDrug

miRDrug: Comprehensive Analysis of the Shared Genes Within miRNA-Drug Pairs Using Grouping, Scoring, and Modeling Approach miRDrug is a novel computational tool aimed at elucidating the complex interactions between microRNAs (miRNAs), diseases, and drugs. Leveraging an innovative Grouping-Scoring-Modeling (G-S-M) approach and integrating data from three biological databases, miRDrug enlightens the genes at the intersection of miRNAs and drug resistance pathways.  This methodology systematically performs the analysis in the following three stages: initially, it groups genes based on shared biological characteristics and their relevance to miRNAs and drug resistance pathways, thereby harnessing the power of collective biological knowledge. Subsequently, it evaluates these groups through a scoring mechanism that quantitatively assesses their potential impact on distinguishing disease-specific phenotypes, particularly focusing on their role in drug resistance. Finally, the modeling phase employs machine learning techniques to construct predictive models based on the highest-scoring gene groups where these genes serve as novel therapeutic targets. This endeavor deepens our understanding of miRNA roles in disease mechanisms. It holds promise for advancing personalized medicine by pinpointing novel therapeutic targets, especially in cancer treatment, where drug resistance remains a formidable challenge. Through meticulous data analysis and validation, miRDrug demonstrates its potential to significantly contribute to advancing targeted treatment strategies, ultimately offering hope for more effective and individualized healthcare solutions. This study underscores the importance of integrating diverse biological datasets to foster a holistic understanding of disease mechanisms, thereby facilitating the discovery of groundbreaking interventions in the fight against complex diseases. Our research has culminated in developing a tool hosted on the Knime platform, showcasing the practical application of our findings using miRDrug. With its user-friendly and modular design, miRDrug invites the broader scientific community to engage with, explore, and further expand our work. miRDrug is available for public access on GitHub, fostering a collaborative space for research enhancement and innovation and inviting enthusiasts and researchers alike to contribute to the ongoing dialogue in the field.

Last edited  Mar 5, 2024 5:14 PM

PriPath Identifying Affected Pathways from Differential Gene Expression via Grouping, Scoring and Modeling with an Embedded Machine Learning Approach

PriPath: Identifying Affected Pathways from Differential Gene Expression via Grouping, Scoring and Modeling with an Embedded Machine Learning Approach Malik Yousef 1, 2, Fatma Ozdemir3, 4, Amhar Jaaber3, Jens Allmer5, and Burcu Bakir-Gungor3 1 Department of Information Systems, Zefat Academic College, Zefat, 13206, Israel 2 Galilee Digital Health Research Center (GDH), Zefat Academic College, Israel 3 Department of Computer Engineering, Faculty of Engineering, Abdullah Gul University, Kayseri, Turkey 4University Institute of Digital Communication Systems, Ruhr-University,Germany 5 Medical Informatics and Bioinformatics, Institute for Measurement Engineering and Sensor Technology, Hochschule Ruhr West, University of Applied Sciences, Mülheim an der Ruhr, Germany * Correspondence: malik.yousef@gmail.com; Abstract Cell homeostasis relies on the concerted actions of several genes; and along this line, dysregulated genes lead to disease manifestations. In living organisms, genes, or their products, do not act alone, but instead act within a large network. Subsets of these networks can be viewed as modules which provide certain functionality for the organism. Kyoto Encyclopedia of Genes and Genomes (KEGG) systematically analyzes gene functions, genes and molecules, and provide a PATHWAY database. Measurements of gene expression (e.g., RNA-seq) can be mapped to KEGG pathways in order to determine which modules are affected in a disease. However, genes acting in multiple pathways, and some other inherent issues complicate such analyses. Current approaches neglect some of the existing knowledge which makes up the KEGG pathways. These approaches may only employ gene expression data to detect dysregulated pathways. However, approaches that take into account more of the compiled information are required for a more holistic association between gene expression and pathways. PriPath is a novel approach that transfers the generic approach of grouping, scoring followed by modeling (G-S-M) for the analysis of gene expression with KEGG pathways. In our approach, we utilize the KEGG pathway as the grouping (term) information and insert this information into a machine learning algorithm for selecting the most significant groups (KEGG pathways). Those groups are utilized to train a machine learning model for the classification task. Our proposed approach successfully assigned KEGG terms to differentially regulated genes with medical relevance. We have tested PriPath on 13 gene expression datasets of various cancers and other diseases. We then compared the performance of PriPath with SVM-RCE, CogNet, and maTE, which are similar in their merit. The results indicate that we outperform maTE in most cases; and PriPath uses less number of genes than SVM-RCE-R and CogNet. For each dataset we manually confirmed the top results in literature and compared PriPath predictions to the predictions of Reactome and DAVID. PriPath can thus aid determining dysregulated pathways, which is applicable to medical diagnostics. In the future, we aim to advance the approach such that it will be possible to perform patient stratification based on gene expression and druggable targets. Thereby, we cover two aspects of precision medicine.

Last edited  Mar 12, 2022 12:58 PM

SEMANT

Last edited  Apr 3, 2024 10:58 PM

SVM-RCE-CW

Last edited  Aug 28, 2024 1:36 PM

TextNetTopics

TextNetTopics: Text Classification based Word Grouping as Topics and Topics' Scoring Malik Yousef1* and Daniel Voskergian2* 1Zefat Academic College, Israel 2Al-Quds University, Palestine Medical document classification is one of the active research problems and the most challenging within the text classification domain. Medical datasets often contain massive feature sets where many features are considered irrelevant, redundant, and add noise, which reduces considerably the classification performance. Thus, to obtain a better accuracy of a classification model, it is crucial to choose a set of features that best discriminate between the classes of medical documents. In this study, we propose TextNetTopics, a novel approach that applies feature selection by considering Bag-of-topics (BOT) rather than the traditional approach, Bag-of-words (BOW). Thus our approach performs topic selections rather than words selection. TextNetTopics is based on the generic approach called G-S-M (Grouping, Scoring, and Modeling), developed by Yousef and his colleagues, where it is used mainly in biological data. The proposed approach suggests scoring topics to select top topics for training the classifier. In this study, we applied TextNetTopics on textual data as a response to the CAMDA challenge. The performance of TextNetTopics outperforms other feature selection approaches while getting a high performance when applying the model on the validation data provided by the CAMDA. Additionally, we have applied our algorithm in different textual datasets.

Last edited  Jun 2, 2022 3:07 PM

TextNetTopics-SFTS-SBTS

Last edited  Mar 17, 2024 11:51 AM

TextNetTopics_TIS

Last edited  Apr 20, 2024 11:00 AM

TextNetTopics_TM

Last edited  Feb 21, 2023 6:51 PM