A subspace learning aided matrix factorization for drug repurposing

سال انتشار: 1402
محل انتشار: دوازدهمین همایش ملی و سومین همایش بین المللی بیوانفورماتیک
کد COI اختصاصی: IBIS12_180
زبان مقاله: انگلیسی
تعداد مشاهده: 183

نویسندگان

Department of Computer Science and Information Technology, Institute for Advanced Studies in BasicSciences (IASBS), Zanjan, Iran

Zahra Narimani

Department of Computer Science and Information Technology, Institute for Advanced Studies in BasicSciences (IASBS), Zanjan, Iran

چکیده

In the rapidly advancing field of drug discovery, the repurposing of existing pharmaceuticalsfor novel therapeutic applications has emerged as a promising strategy. Traditionally, drug design hasbeen a costly and time-consuming endeavor, but recent advancements in high-throughput technologiesand machine learning have significantly streamlined the process. This paper presents a novel approachto drug repurposing that integrates state-of-the-art computational methods, including graph-basedanalysis, matrix factorization, and machine learning techniques, to enhance the predictive capabilitiesof an existing model, iDrug, which was based on matrix factorization. The iDrug model, a pioneeringtool in drug repositioning, operates within the interconnected domains of drug-disease and drug-targetnetworks. By harnessing the power of graph-based methods, we can analyze the intricate relationshipsbetween drugs, targets, side-effects, and diseases, uncovering patterns that may not be apparent throughtraditional methods. Matrix factorization techniques are employed to decompose sparse drug-target anddrug-disease matrices, revealing latent features that can predict drug-target and drug-disease similaritieswith unprecedented accuracy. To further refine the iDrug model, we propose the integration of Sparseand Low-Redundant Subspace Learning-based DualGraph Regularized Robust Feature Selection(SLSDR). SLSDR is an efficient method that addresses the challenges of feature selection in highdimensionaldatasets, extracting meaningful patterns while discarding irrelevant or redundant features.This integration enhances the model's predictive accuracy, interpretability, and scalability, making it amore powerful and versatile tool for drug discovery. By leveraging the strengths of both computationalmethods and machine learning techniques, we have developed a novel framework that not only expandsthe potential therapeutic applications of drugs but also represents a significant step forward in the field,paving the way for more efficient and effective drug discovery processes.

کلیدواژه ها

Bioinformatics; Drug Repositioning; Machine Learning; Computational Biology; SLSDR; iDrug

مقالات مرتبط جدید

اطلاعات بیشتر در مورد COI

COI مخفف عبارت CIVILICA Object Identifier به معنی شناسه سیویلیکا برای اسناد است. COI کدی است که مطابق محل انتشار، به مقالات کنفرانسها و ژورنالهای داخل کشور به هنگام نمایه سازی بر روی پایگاه استنادی سیویلیکا اختصاص می یابد.

کد COI به مفهوم کد ملی اسناد نمایه شده در سیویلیکا است و کدی یکتا و ثابت است و به همین دلیل همواره قابلیت استناد و پیگیری دارد.