2.3 Calculate Gaussian Interaction Profile Kernel Similarity

TD Tao Duan
ZK Zhufang Kuang
JW Jiaqi Wang
ZM Zhihao Ma
request Request a Protocol
ask Ask a question
Favorite

In this section, the adjacency matrix A is constructed according to the known lncRNA–disease association. The A (l i , d j ) indicates whether lncRNA l i and disease d j are related. The A as shown in formula (6):

In lncRNA functional similarity matrix FSL and the disease semantic similarity matrix SSD, many elements are 0. In order to make the similarity network not sparse, Gaussian kernel similarity is calculated, as shown in formula (7):

where GSL (l m , l n ) is the Gaussian interaction profile kernel similarity score of lncRNA l m and l n , and GSD (d i , d j ) is the Gaussian interaction profile kernel similarity of diseases d i and d j . The A (m,) is mth row of A, and A (. i) is ith col of A. Parameters δ l and δ d are obtained as shown in Eq. 9 and Eq. 10:

where p is the number of lncRNAs and q is the number of diseases.

Do you have any questions about this protocol?

Post your question to gather feedback from the community. We will also invite the authors of this article to respond.

post Post a Question
0 Q&A