Mukund Kavitha, Subramaniam Shankar
Department of Bioengineering, University of California, San Diego, La Jolla, CA, United States.
Departments Cellular and Molecular Medicine, Computer Science and Engineering, University of California, San Diego, La Jolla, CA, United States.
Front Physiol. 2017 Dec 1;8:980. doi: 10.3389/fphys.2017.00980. eCollection 2017.
Diseases affecting skeletal muscle exhibit considerable heterogeneity in intensity, etiology, phenotypic manifestation and gene expression. Systems biology approaches using network theory, allows for a holistic understanding of functional similarities amongst diseases. Here we propose a co-expression based, network theoretic approach to extract functional similarities from 20 heterogeneous diseases comprising of dystrophinopathies, inflammatory myopathies, neuromuscular, and muscle metabolic diseases. Utilizing this framework we identified seven closely associated disease clusters with 20 disease pairs exhibiting significant correlation ( < 0.05). Mapping the diseases onto a human protein-protein interaction network enabled the inference of a common program of regulation underlying more than half the muscle diseases considered here and referred to as the "protein signature." Enrichment analysis of 17 protein modules identified as part of this signature revealed a statistically non-random dysregulation of muscle bioenergetic pathways and calcium homeostasis. Further, analysis of mechanistic similarities of less explored significant disease associations [such as between amyotrophic lateral sclerosis (ALS) and cerebral palsy (CP)] using a proposed "functional module" framework revealed adaptation of the calcium signaling machinery. Integrating drug-gene information into the quantitative framework highlighted the presence of therapeutic opportunities through drug repurposing for diseases affecting the skeletal muscle.
影响骨骼肌的疾病在强度、病因、表型表现和基因表达方面存在很大的异质性。使用网络理论的系统生物学方法能够全面理解疾病之间的功能相似性。在此，我们提出一种基于共表达的网络理论方法，从包括肌营养不良症、炎性肌病、神经肌肉疾病和肌肉代谢疾病在内的20种异质性疾病中提取功能相似性。利用这个框架，我们确定了七个紧密相关的疾病簇，其中20对疾病表现出显著相关性（<0.05）。将这些疾病映射到人类蛋白质-蛋白质相互作用网络上，能够推断出这里所考虑的一半以上肌肉疾病背后的共同调控程序，即“蛋白质特征”。对确定为该特征一部分的17个蛋白质模块进行的富集分析显示，肌肉生物能途径和钙稳态存在统计学上非随机的失调。此外，使用提出的“功能模块”框架对较少探索的显著疾病关联（如肌萎缩侧索硬化症（ALS）和脑瘫（CP）之间）的机制相似性进行分析，揭示了钙信号传导机制的适应性变化。将药物-基因信息整合到定量框架中，突出了通过药物重新利用来治疗影响骨骼肌疾病的治疗机会。
