IRWRLDA, for lncRNA-disease interaction prediction. The work
constructed an lncRNA similarity network by integrating known
lncRNA-disease associations, disease semantic similarity, and vari-
ous lncRNA similarity measures. The novelty of IRWRLDA lies in
the incorporation of lncRNA expression similarity and disease
semantic similarity to set the initial probability vector of the
RWR. Therefore, IRWRLDA could be applied to diseases without
any known related lncRNAs. In addition, Zhou et al. [47] con-
structed an lncRNA-lncRNA crosstalk network by examining the
significant co-occurrence of shared miRNA response elements on
lncRNA transcripts from the competing endogenous RNAs view-
point. As expected, functional analysis showed that lncRNAs shar-
ing significantly enriched interacting miRNAs tend to be involved
in similar diseases and have more functionally related flanking gene
sets. They further proposed a novel rank-based method,
RWRHLD, to prioritize candidate lncRNA-disease associations by
integrating three networks (miRNA-associated lncRNA-lncRNA
crosstalk network, disease-disease similarity network, and known
lncRNA-disease association network) into a heterogeneous net-
work and implementing a random walk with restart on this hetero-
geneous network. Based on knowledge that lncRNAs have
relationship with disease by regulating the expression of disease
gene, Alaimo et al. [48] presented an information propagation
method, ncPred, for novel ncRNA-disease association inference.
In the first step, they constructed a tripartite network based on two
levels of interaction: ncRNA-target and target-disease. Then, the
network-based inference method was employed to predict potential
lncRNA-disease associations.
Considering the difficulty of lncRNA similarity network con-
struction, some researchers try to construct multi-level network of
lncRNA-disease. Then, new disease-related lncRNAs are identified
based on multi-level network.
Liu et al. [49] developed a computational method to identify
candidate cancer-related lncRNAs based on interactions between
protein-coding genes (PCGs) and lncRNAs. They constructed the
lncRNA-PCG bipartite network of prostate cancer by combining
expression profiles of lncRNAs and PCGs and protein-protein
interactions. Six prostate cancer-related lncRNAs were included in
this bipartite network. Based on this network, the random walk
method was utilized to identify lncRNA related with prostate can-
cer. This method found that lncRNA ENSG00000261777 shares
an intron with DDX19 and interacts with IGF2 P1, indicating its
involvement in prostate cancer. In consideration that phenotype
data directly reflected disease association, Yao et al. [50] proposed a
novel algorithm, LncPriCNet, to prioritize candidate lncRNAs
associated with diseases based on a multi-level composite network.
They constructed a composite network by combining phenotype-
phenotype interactions, lncRNA-lncRNA interactions, and gene-Identifying Interactions Between Long Noncoding RNAs and Diseases... 211