Abstract
MicroRNAs (miRNA) are known to be involved in the development of various diseases. Hence various scientists in the field have been utilized computational analyses to determine the relationship between miRNA and diseases. However, the knowledge of miRNA and disease is still very limited. Therefore, we combined Environmental Factor (EF) data to a miRNA global network. Increasing research has shown that relationship between miRNAs and EFs play a significant role in classifying types of diseases. Environmental Factors consist of radiation, drugs, viruses, alcohol, cigarettes, and stress. Our global network considered all the interactions between every pair of miRNAs, which has led to precise analyses in comparison to local networks. As a result, our approaches’ performance demonstrated its effectiveness in identifying disease-related miRNA and this is the area under the ROC curve (AUC) of 74.46%. Furthermore, comparative experiment has shown that our approach performs comparable to other existing methods with an accuracy of 94%, 90% and 96% for breast cancer, colonic cancer, and lung cancer respectively. In conclusion, these results support that our research has broadened new biological insights on identifying disease-related miRNAs.