Currently, the research of multi-omics, such as genomics, proteinomics, transcriptomics, microbiome, metabolomics, pathomics, and radiomics, are hot spots. The relationship between multi-omics data, drugs, and diseases has received extensive attention from researchers. At the same time, multi-omics can effectively predict the diagnosis, prognosis, and treatment of diseases. In essence, these research entities, such as genes, RNAs, proteins, microbes, metabolites, pathways as well as pathological and medical imaging data, can all be represented by the network at different levels. And some computer and biology scholars have tried to use computational methods to explore the potential relationships between biological entities. We summary a comprehensive research strategy, that is to build a multi-omics heterogeneous network, covering multimodal data, and use the current popular computational methods to make predictions. In this study, we first introduce the calculation method of the similarity of biological entities at the data level, second discuss multimodal data fusion and methods of feature extraction. Finally, the challenges and opportunities at this stage are summarized. Some scholars have used such a framework to calculate and predict. We also summarize them and discuss the challenges. We hope that our review could help scholars who are interested in the field of bioinformatics, biomedical image, and computer research.

中文翻译：

---

基因组学、蛋白质组学、转录组学、微生物组学、代谢组学、病理组学、放射组学、药物、症状、环境因素和疾病网络的关联预测：综合方法

目前，基因组学、蛋白质组学、转录组学、微生物组学、代谢组学、病理组学、放射组学等多组学研究是热点。多组学数据、药物和疾病之间的关系受到了研究人员的广泛关注。同时，多组学可以有效预测疾病的诊断、预后和治疗。从本质上讲，这些研究实体，如基因、RNA、蛋白质、微生物、代谢物、通路以及病理和医学影像数据，都可以通过网络在不同层次上进行表示。并且一些计算机和生物学学者已经尝试使用计算方法来探索生物实体之间的潜在关系。我们总结了一个综合研究策略，即构建多组学异构网络，覆盖多模态数据，并使用当前流行的计算方法进行预测。在本研究中，我们首先在数据层面介绍了生物实体相似度的计算方法，然后讨论了多模态数据融合和特征提取方法。最后总结了现阶段面临的挑战和机遇。一些学者使用这样的框架进行计算和预测。我们还总结了它们并讨论了挑战。我们希望我们的评论可以帮助对生物信息学、生物医学图像和计算机研究领域感兴趣的学者。总结了现阶段的挑战和机遇。一些学者使用这样的框架进行计算和预测。我们还总结了它们并讨论了挑战。我们希望我们的评论可以帮助对生物信息学、生物医学图像和计算机研究领域感兴趣的学者。总结了现阶段的挑战和机遇。一些学者使用这样的框架进行计算和预测。我们还总结了它们并讨论了挑战。我们希望我们的评论可以帮助对生物信息学、生物医学图像和计算机研究领域感兴趣的学者。