Supporting data for "Bioentity2vec: Attribute- and Behavior-driven Representation for Multi-type Relationship Prediction between Various Bioentities"

The explosive growth of genomic, chemical and pathological data provides new opportunities and challenges for humans to reexamine life activities in cells. However, there exist few computational models that aggregate various bioentities to comprehensively reveal the physical and functional landscape of biological system. Here, we construct a graph called Molecular Association Network (MAN) and a representation method called Bioentity2vec. Specifically, MAN is a heterogeneous attribute network consists of 18 kinds of edges (relationships) among 8 kinds of nodes (bioentities). Bioentity2vec is an algorithm that represents the nodes as vectors by integrating bioentity attribute such as RNA sequence and bioentity behavior that is the relationship between bioentities. Then, random forest classifier is applied to carry out the relationship prediction task. The proposed approach achieved promising performance on 18 relationships, with AUC of 0.9608 and AUPR of 0.9572. The results strongly prove that MAN is a network with rich topological and biological information and Bioentity2vec can adequately characterize bioentities. Generally, our method can achieve simultaneous prediction of both single-type and multi-type relationships, which bring beneficial inspiration to relevant scholars and expand the medical research paradigm.