Many complex diseases occur due to genetic factors. A perturbation in the pathway of gene interactions leads to such disorders. Even though a group of genes is responsible, a few significant genes act as a biomarker for disease, perturbing the healthy network. Identifying such marker genes or a set of genes that play a pivotal role in diseases helps drug prioritization.

We propose a scheme for finding potential bio-markers using a multi-layer consensus-driven approach. We reconstruct a functional module guided disease sub-network, followed by a multi-step consensus of network inference methods and shared ontological terms. We perform centrality analysis on the sub-networks under consideration and report hub genes as potentially key players in the target disease.

To establish our scheme's effectiveness, we use Alzheimer's Disease (AD) and Breast Cancer as candidate diseases for experimentation. We evaluate the significance of prioritized genes based on reported evidence. We observe that BRCA1, BRCA2, and PTEN are the essential genes for Breast Cancer, whereas MAPK1, APP, and CASP7 are the essential genes playing an important role during AD.

由于遗传因素，会发生许多复杂的疾病。基因相互作用途径中的扰动导致此类疾病。即使一组基因负责，一些重要的基因还是疾病的生物标志物，扰乱了健康的网络。鉴定出这些标记基因或一组在疾病中起关键作用的基因有助于药物优先排序。

我们提出了一种使用多层共识驱动方法寻找潜在生物标志物的方案。我们重建了一个功能模块指导的疾病子网络，然后是网络推理方法和共享的本体论术语的多步骤共识。我们对所考虑的子网络进行集中性分析，并将中心基因报告为目标疾病的潜在关键参与者。

为了确定我们计划的有效性，我们将阿尔茨海默氏病（AD）和乳腺癌作为实验的候选疾病。我们根据报告的证据评估优先基因的重要性。我们观察到BRCA1，BRCA2和PTEN是乳腺癌的必需基因，而MAPK1，APP和CASP7是在AD中起重要作用的必需基因。