Reverter Antonio, Fortes Marina R S
CSIRO Livestock Industries, Queensland Bioscience Precinct, Brisbane, QLD, Australia.
Methods Mol Biol. 2013;1019:437-47. doi: 10.1007/978-1-62703-447-0_20.
In this chapter we describe the Association Weight Matrix (AWM), a novel procedure to exploit the results from genome-wide association studies (GWAS) and, in combination with network inference algorithms, generate gene networks with regulatory and functional significance. In simple terms, the AWM is a matrix with rows represented by genes and columns represented by phenotypes. Individual {i, j}th elements in the AWM correspond to the association of the SNP in the ith gene to the jth phenotype. While our main objective is to provide a recipe-like tutorial on how to build and use AWM, we also take the opportunity to briefly reason the logic behind each step in the process. To conclude, we discuss the impact on AWM of issues like the number of phenotypes under scrutiny, the density of the SNP chip and the choice of contrast upon which to infer the cause-effect regulatory interactions.
在本章中，我们描述了关联权重矩阵（AWM），这是一种利用全基因组关联研究（GWAS）结果的新方法，并结合网络推理算法，生成具有调控和功能意义的基因网络。简单来说，AWM是一个矩阵，其行由基因表示，列由表型表示。AWM中第{i, j}个元素对应于第i个基因中的单核苷酸多态性（SNP）与第j个表型的关联。虽然我们的主要目标是提供一个类似食谱的教程，介绍如何构建和使用AWM，但我们也借此机会简要阐述该过程中每一步背后的逻辑。最后，我们讨论了诸如受审查的表型数量、SNP芯片的密度以及用于推断因果调控相互作用的对比选择等问题对AWM的影响。
