The phenotype-genotype relationship is a key for personalized and precision medicine for complex diseases. To unravel the complexity of the clinical phenotype-genotype network, we used cardiovascular diseases (CVDs) and associated non-coding RNAs (ncRNAs) (i.e. miRNAs, long ncRNAs, etc.) as the case for the study of CVDs at a systems or network level. We first integrated a database of CVDs and ncRNAs (CVDncR, http://sysbio.org.cn/cvdncr/) to construct CVD-ncRNA networks and annotate their clinical associations. To characterize the networks, we then separated the miRNAs into two groups, i.e. universal miRNAs associated with at least two types of CVDs and specific miRNAs related only to one type of CVD. Our analyses indicated two interesting patterns in these CVD-ncRNA networks. First, scale-free features were present within both CVD-miRNA and CVD-lncRNA networks; second, universal miRNAs were more likely to be CVDs biomarkers. These results were confirmed by computational functional analyses. The findings offer theoretical guidance for decoding CVD-ncRNA associations and will facilitate the screening of CVD ncRNA biomarkers. Database URL: http://sysbio.org.cn/cvdncr/.
© The Author(s) 2020. Published by Oxford University Press.