School of Computer Science and Technology, Xidian University, Xi'an 710071, China.
College of Information Engineering, Northwest Agriculture and Forestry University, Yangling, Xianyang 712100, China.
Molecules. 2017 Dec 14;22(12):2228. doi: 10.3390/molecules22122228.
The advances in biological technologies make it possible to generate data for multiple conditions simultaneously. Discovering the condition-specific modules in multiple networks has great merit in understanding the underlying molecular mechanisms of cells. The available algorithms transform the multiple networks into a single objective optimization problem, which is criticized for its low accuracy. To address this issue, a multi-objective genetic algorithm for condition-specific modules in multiple networks (MOGA-CSM) is developed to discover the condition-specific modules. By using the artificial networks, we demonstrate that the MOGA-CSM outperforms state-of-the-art methods in terms of accuracy. Furthermore, MOGA-CSM discovers stage-specific modules in breast cancer networks based on The Cancer Genome Atlas (TCGA) data, and these modules serve as biomarkers to predict stages of breast cancer. The proposed model and algorithm provide an effective way to analyze multiple networks.
生物技术的进步使得同时生成多种条件的数据成为可能。在多个网络中发现特定于条件的模块，对于理解细胞的潜在分子机制具有重要意义。现有的算法将多个网络转化为一个单一的目标优化问题，这种方法因其准确性低而受到批评。为了解决这个问题，开发了一种用于多个网络中特定于条件的模块的多目标遗传算法（MOGA-CSM），以发现特定于条件的模块。通过使用人工网络，我们证明 MOGA-CSM 在准确性方面优于最先进的方法。此外，MOGA-CSM 基于癌症基因组图谱（TCGA）数据发现乳腺癌网络中的阶段特异性模块，这些模块可用作预测乳腺癌阶段的生物标志物。所提出的模型和算法为分析多个网络提供了一种有效的方法。
