Inadequate dietary fiber consumption has become common across industrialized nations, accompanied by changes in gut microbial composition and a dramatic increase in chronic metabolic diseases. The human gut microbiome harbors genes that are required for the digestion of fiber, resulting in the production of end products that mediate gastrointestinal and systemic benefits to the host. Thus, the use of fiber interventions has attracted increasing interest as a strategy to modulate the gut microbiome and improve human health. However, considerable interindividual differences in gut microbial composition have resulted in variable responses toward fiber interventions. This variability has led to observed nonresponder individuals and highlights the need for personalized approaches to effectively redirect the gut ecosystem. In this review, we summarize strategies used to address the responder and nonresponder phenomenon in dietary fiber interventions and propose a targeted approach to identify predictive features based on knowledge of fiber metabolism and machine learning approaches.

中文翻译：

---

预测对膳食纤维干预的个性化反应：调节肠道微生物组以改善健康的机会

膳食纤维摄入不足在工业化国家已变得普遍，伴随着肠道微生物组成的变化和慢性代谢疾病的急剧增加。人类肠道微生物组含有消化纤维所需的基因，从而产生最终产物，为宿主带来胃肠道和全身益处。因此，使用纤维干预作为调节肠道微生物组和改善人类健康的策略引起了越来越多的兴趣。然而，肠道微生物组成的巨大个体差异导致了对纤维干预的不同反应。这种变异性导致观察到无反应个体，并强调需要个性化方法来有效地重新引导肠道生态系统。在这篇综述中，我们总结了用于解决膳食纤维干预中的反应者和无反应者现象的策略，并提出了一种基于纤维代谢知识和机器学习方法来识别预测特征的有针对性的方法。