Identifying miRNA synergism using multiple-intervention causal inference

Junpeng Zhang, Vu Viet Hoang Pham, Lin Liu, Taosheng Xu, Buu Minh Thanh Truong, Jiuyong Li, Nini Rao, Thuc Duy Le

Abstract

The miRsyn framework introduces a novel computational approach to identify how multiple microRNAs (miRNAs) work together to regulate genes through "multiple-intervention" causal inference. By simulating the simultaneous knockdown of several miRNAs using observational data, the researchers identified a biologically meaningful, small-world network of miRNA interactions critical to breast cancer. miRsyn outperforms traditional single-intervention methods by capturing the joint regulatory effects that drive complex disease outcomes.

Introduction

MicroRNAs are short RNA molecules that regulate gene expression, often acting as "dimmer switches" for protein production. While it is established that miRNAs often work synergistically to regulate complex diseases like cancer, identifying these functional teams remains a significant challenge. Most existing methods suffer from two primary limitations: a reliance on static sequence-binding data, which does not reflect real-time biological activity, and a focus on single-intervention causal models. These models simulate "knocking down" only one miRNA at a time, failing to capture the true nature of synergism where the combined effect of multiple miRNAs drives regulatory outcomes.

Methods

The authors developed miRsyn, a step-wise computational pipeline designed to overcome these limitations. The framework begins with Feature Selection using a Cox regression model to identify miRNAs and mRNAs most significant to breast cancer. The core of the pipeline is Multiple-Intervention Causal Inference, utilizing the Joint-IDA algorithm to estimate the joint causal effect of a set of miRNAs on a target gene. This process mathematically simulates the simultaneous silencing of multiple miRNAs to observe changes in target gene expression.

Overview of the miRsyn computational pipeline for identifying miRNA synergistic modules.

The causal scores are integrated with validated sequence-binding data from miRTarBase to ensure physical plausibility. Finally, bi-clique algorithms are used to discover functional modules, which are then subjected to Gene Ontology (GO) and KEGG pathway enrichment analysis. The study utilized matched miRNA and mRNA expression data from The Cancer Genome Atlas (TCGA), focusing on 753 breast cancer (BRCA) samples.

Results and Key Insights

The miRsyn framework identified 702 synergistic miRNA-miRNA pairs, a substantial improvement over the 239 pairs found by the previous state-of-the-art "single-intervention" method, mirSRN. The identified synergistic modules were significantly enriched in critical cancer pathways, including cell cycle arrest, apoptosis, and the VEGF signaling pathway.

The identified miRNA synergistic network in breast cancer, exhibiting small-world architecture and hub miRNAs.

Key findings include:

  • Sequence vs. Synergy: Many miRNA pairs sharing sequence-level targets do not work together at the expression level, highlighting the necessity of observational expression data.
  • Small-World Architecture: The miRNA synergistic network exhibits a "small-world" topology, characterized by a high clustering coefficient (0.528) and short path lengths. This suggests a highly efficient and robust regulatory system.
  • Hub miRNAs: The most connected nodes in the network (hubs) were identified as known breast cancer-related miRNAs, such as miR-10a-5p and miR-26a-5p, validating the model's biological relevance.
  • Co-expression: Approximately 71% of synergistic pairs showed similar expression patterns, suggesting that miRNAs working together are often co-regulated to ensure rapid cellular responses.

Conclusion

miRsyn represents a significant advancement in identifying miRNA synergism by moving beyond single-intervention models. While the method is computationally intensive and relies on the "faithfulness" assumption of causal inference, it successfully identifies biologically relevant regulatory modules in breast cancer. Future work may involve incorporating time-series data to capture the dynamic evolution of these synergistic relationships during cancer progression.

Summary updated

Abstract

Abstract Background Studying multiple microRNAs (miRNAs) synergism in gene regulation could help to understand the regulatory mechanisms of complicated human diseases caused by miRNAs. Several existing methods have been presented to infer miRNA synergism. Most of the current methods assume that miRNAs with shared targets at the sequence level are working synergistically. However, it is unclear if miRNAs with shared targets are working in concert to regulate the targets or they individually regulate the targets at different time points or different biological processes. A standard method to test the synergistic activities is to knock-down multiple miRNAs at the same time and measure the changes in the target genes. However, this approach may not be practical as we would have too many sets of miRNAs to test. Results In this paper, we present a novel framework called miRsyn for inferring miRNA synergism by using a causal inference method that mimics the multiple-intervention experiments, e.g. knocking-down multiple miRNAs, with observational data. Our results show that several miRNA-miRNA pairs that have shared targets at the sequence level are not working synergistically at the expression level. Moreover, the identified miRNA synergistic network is small-world and biologically meaningful, and a number of miRNA synergistic modules are significantly enriched in breast cancer. Our further analyses also reveal that most of synergistic miRNA-miRNA pairs show the same expression patterns. The comparison results indicate that the proposed multiple-intervention causal inference method performs better than the single-intervention causal inference method in identifying miRNA synergistic network. Conclusions Taken together, the results imply that miRsyn is a promising framework for identifying miRNA synergism, and it could enhance the understanding of miRNA synergism in breast cancer.
BibTeX
@article{zhang2019identifying,
  author = {Junpeng Zhang and Vu Viet Hoang Pham and Lin Liu and Taosheng Xu and Buu Minh Thanh Truong and Jiuyong Li and Nini Rao and Thuc Duy Le},
  title = {{Identifying miRNA synergism using multiple-intervention causal inference}},
  journal = {BMC Bioinformatics},
  year = {2019},
  doi = {10.1186/s12859-019-3215-5},
}
Figure