A message passing framework with multiple data integration for miRNA-disease association prediction

Abstract

Micro RNA or miRNA is a highly conserved class of non-coding RNA that plays an important role in many diseases. Identifying miRNA-disease associations can pave the way for better clinical diagnosis and finding potential drug targets. We propose a biologically-motivated data-driven approach for the miRNA-disease association prediction, which overcomes the data scarcity problem by exploiting information from multiple data sources. The key idea is to enrich the existing miRNA/disease-protein-coding gene (PCG) associations via a message passing framework, followed by the use of disease ontology information for further feature filtering. The enriched and filtered PCG associations are then used to construct the inter-connected miRNA-PCG-disease network to train a structural deep network embedding (SDNE) model. Finally, the pre-trained embeddings and the biologically relevant features from the miRNA family and disease semantic similarity are concatenated to form the pair input representations to a Random Forest classifier whose task is to predict the miRNA-disease association probabilities. We present large-scale comparative experiments, ablation, and case studies to showcase our approach's superiority. Besides, we make the model prediction results for 1618 miRNAs and 3679 diseases, along with all related information, publicly available at http://software.mpm.leibniz-ai-lab.de/ to foster assessments and future adoption.

Figure 1

The Kaplan survival curve of PBLL patients.

Figure 2

Kaplan–Meyer survival curves of PBLL patients stratified by the top miRNAs with the top highest prediction scores.

Figure 3

MPM’s architecture. MPM consists of a message passing layer (section “The message passing framework/module”) , a feature selection with a side supervised task (section “The feature selection module”), a Structural Deep Embedding network (section “The structural embedding learning”), and a binary classifier (section “The classification module”).

Figure 4

An example of the protein functional interaction network with the various relation types highlighted by different colors.

Figure 5

An example of how a message passing framework functions. The numbers inside the circles indicate nodes’ IDs. ‘w’ indicates the node feature weight (as described in section “The message passing framework/module”). In the first iteration, new weights for nodes 4, 6, 7 are calculated according to equation (1). Only the weight for node 6 gets updated during the second iteration.

Figure 6

The final miRNA-disease input pair representation.