Input: Matrices Y n × m, MFSn × n, DSSm × m, parameter α, β

Output: Predicted associations matrix F.

1: For i ← 1 to n do

2: Select the sparse neighborhood for miRNA i by Equation (5).

3: End for

4: Obtain a reconstructed miRNA similarity matrix by Equation (6).

5: For j ← 1 to m do

6: Select the sparse neighborhood for disease j by Equation (5).

7: End for

8: Obtain a reconstructed disease similarity matrix by Equation (6).

9: Integrate similarity information by Equations (7) and (8).

10: Repeat

11: $F_M\left(t+1\right)=\alpha \ast MS\ast F_M\left(t\right)+\left(1-\alpha \right)\ast Y$

12: Until convergence

13: Repeat

14: $F_D\left(t+1\right)=\alpha \ast DS\ast F_D\left(t\right)+\left(1-\alpha \right)\ast Y^{\prime }$

15: Until convergence

16: $F=\beta F_M+\left(1-\beta \right){F_D}^{\prime }$

17: Return F