Discovering causal interactions using Bayesian network scoring and information gain.
Discovering causal interactions using Bayesian network scoring and information gain.
复制标题
DOI：
10.1186/s12859-016-1084-8
复制
发表时间：
2016-05-26
影响因子：
3
通讯作者：
Neapolitan R
中科院分区：
文献类型：
--
作者：
Zeng Z;Jiang X;Neapolitan R
从数据中学习因果影响的问题最近引起了广泛关注。标准统计方法可能难以学习离散原因，这些原因相互作用以影响目标，因为这些方法中的假设通常不能很好地模拟离散因果关系。那么一项重要的任务就是从数据中学习这种交互。受从全基因组关联研究（GWAS）开发的数据集中学习上位相互作用问题的启发，研究人员构想了学习离散相互作用的新方法。然而，这些方法中的许多方法并没有区分代表真实交互的模型和代表具有强烈个体影响的非交互原因的模型。最近的算法 MBS-IGain 通过使用贝叶斯网络学习和信息增益来发现高维数据集中的交互，从而解决了这一难题。然而，MBS-IGain 需要边际效应来检测包含两个以上原因的相互作用。如果数据集不是高维的，我们可以通过穷举搜索来避免这个缺点。我们开发了 Exhaustive-IGain，它类似于 MBS-IGain，但进行了详尽的搜索。我们使用基于边际效应相互作用的低维模拟数据集和基于无边际效应相互作用的低维模拟数据集来比较 Exhaustive-IGain 与 MBS-IGain 的性能。它们在基于边际效应的数据集上的表现相似。然而，Exhaustive-IGain 在基于无边际效应的 3 因和 4 因交互作用的数据集上明显优于 MBS-IGain。我们应用 Exhaustive-IGain 来研究临床变量如何相互作用影响乳腺癌生存，并获得与乳腺癌肿瘤学家的判断一致的结果。我们得出的结论是，如果我们执行详尽的搜索，信息增益和贝叶斯网络评分的结合使用使我们能够发现没有边际效应的高阶相互作用。我们进一步得出结论，Exhaustive-IGain 在应用于实际数据时可以是有效的。本文的在线版本 (doi:10.1186/s12859-016-1084-8) 包含补充材料，可供授权用户使用。
The problem of learning causal influences from data has recently attracted much attention. Standard statistical methods can have difficulty learning discrete causes, which interacting to affect a target, because the assumptions in these methods often do not model discrete causal relationships well. An important task then is to learn such interactions from data. Motivated by the problem of learning epistatic interactions from datasets developed in genome-wide association studies (GWAS), researchers conceived new methods for learning discrete interactions. However, many of these methods do not differentiate a model representing a true interaction from a model representing non-interacting causes with strong individual affects. The recent algorithm MBS-IGain addresses this difficulty by using Bayesian network learning and information gain to discover interactions from high-dimensional datasets. However, MBS-IGain requires marginal effects to detect interactions containing more than two causes. If the dataset is not high-dimensional, we can avoid this shortcoming by doing an exhaustive search. We develop Exhaustive-IGain, which is like MBS-IGain but does an exhaustive search. We compare the performance of Exhaustive-IGain to MBS-IGain using low-dimensional simulated datasets based on interactions with marginal effects and ones based on interactions without marginal effects. Their performance is similar on the datasets based on marginal effects. However, Exhaustive-IGain compellingly outperforms MBS-IGain on the datasets based on 3 and 4-cause interactions without marginal effects. We apply Exhaustive-IGain to investigate how clinical variables interact to affect breast cancer survival, and obtain results that agree with judgements of a breast cancer oncologist. We conclude that the combined use of information gain and Bayesian network scoring enables us to discover higher order interactions with no marginal effects if we perform an exhaustive search. We further conclude that Exhaustive-IGain can be effective when applied to real data. The online version of this article (doi:10.1186/s12859-016-1084-8) contains supplementary material, which is available to authorized users.