Haerianardakani Amirhossein, Taheri Golnaz
Department of Computer and Systems Sciences, Stockholm University, Stockholm 16455, Sweden.
School of Electrical Engineering and Computer Science, SciLifeLab, KTH Royal Institute of Technology, Stockholm 10044, Sweden.
Bioinform Adv. 2025 Apr 29;5(1):vbaf094. doi: 10.1093/bioadv/vbaf094. eCollection 2025.
We propose a network-based unsupervised learning model to identify essential cancer genes and modules for 12 different cancer types, supported by a Python package for practical application. The model constructs a gene network from frequently mutated genes and biological processes, ranks genes using topological features, and detects critical modules. Evaluation across cancer types confirms its effectiveness in prioritizing cancer-related genes and uncovering relevant modules. The Python package allows users to input gene lists, retrieve rankings, and identify associated modules. This work provides a robust method for gene prioritization and module detection, along with a user-friendly package to support research and clinical decision-making in cancer genomics.
GenePioneer is released as an open-source software under the MIT license. The source code is available on GitHub at https://github.com/Golnazthr/ModuleDetection.
我们提出了一种基于网络的无监督学习模型，用于识别12种不同癌症类型的关键癌症基因和模块，并提供了一个用于实际应用的Python包。该模型从频繁突变的基因和生物学过程构建基因网络，利用拓扑特征对基因进行排名，并检测关键模块。跨癌症类型的评估证实了其在对癌症相关基因进行优先级排序和发现相关模块方面的有效性。该Python包允许用户输入基因列表、检索排名并识别相关模块。这项工作为基因优先级排序和模块检测提供了一种强大的方法，以及一个用户友好的包，以支持癌症基因组学的研究和临床决策。
GenePioneer作为开源软件根据MIT许可发布。源代码可在GitHub上获取，网址为https://github.com/Golnazthr/ModuleDetection 。
