Until date, several machine learning approaches have been proposed for the dynamic modeling of temporal omics data. Although they have yielded impressive results in terms of model accuracy and predictive ability, most of these applications are based on "Black-box" algorithms and more interpretable models have been claimed by the research community. The recent eXplainable Artificial Intelligence (XAI) revolution offers a solution for this issue, were rule-based approaches are highly suitable for explanatory purposes. The further integration of the data mining process along with functional-annotation and pathway analyses is an additional way towards more explanatory and biologically soundness models. In this paper, we present a novel rule-based XAI strategy (including pre-processing, knowledgeextraction and functional validation) for finding biologically relevant sequential patterns from longitudinal human gene expression data (GED). To illustrate the performance of our pipeline, we work on in vivo temporal GED collected within the course of a long-term dietary intervention in 57 subjects with obesity (GSE77962). As validation populations, we employ three independent datasets following the same experimental design. As a result, we validate primarily extracted gene patterns and prove the goodness of our strategy for the mining of biologically relevant gene-gene temporal relations. Our whole pipeline has been gathered under open-source software and could be easily extended to other human temporal GED applications.

Association rules are commonly used to provide decision‐makers with knowledge that helps them to make good decisions. Most of the published proposals mine association rules without paying particular attention to temporal information. However, in real‐life applications data usually change over time or presenting different temporal situations. Therefore, the extracted knowledge may not be useful, since we may not know whether the rules are currently applicable or whether they will be applicable in the future. For this reason, in recent years, many methods have been proposed in the literature for mining temporal association rules, which introduce a greater predictive and descriptive power providing an additional degree of interestingness. One of the main problems in this research field is the lack of visibility most works suffer since there is no standard terminology to refer to it, making it difficult to find and compare proposals and studies in the field. This contribution attempts to offer a well‐defined framework that allows researchers both to easily locate the previous proposals and to propose well‐grounded methods in the future. To accomplish both objectives, a two‐level taxonomy is proposed according to whether the time variable is considered to provide order to the data collection and to locate some temporal constraints, or whether it is considered as an attribute within the learning process. Some recent applications, available software tools, and a bibliographical analysis in accordance with the Web of Science are also shown. Finally, some critical considerations and potential further directions are discussed. This article is categorized under: Technologies > Association Rules Algorithmic Development > Association Rules

No abstract

No abstract