Algorithm : The algorithm of our proposed method |
---|
Input: Known associations Ytrain∈Rn×m, disease space kernels (\(\mathbf {K}^{d}_{SEM},\mathbf {K}^{d}_{GIP}\in \mathbf {R}^{n\times n}\)) and gene space kernels (\(\mathbf {K}^{g}_{GO},\mathbf {K}^{g}_{PPI},\mathbf {K}^{g}_{SW},\mathbf {K}^{g}_{GIP}\in \mathbf {R}^{m\times m}\)), parameters λd,λg,β and k-Nearest Neighbor for DHRLS; |
Output: Predicted associations F∗∈Rn×m; |
1.Calculating disease and gene kernels, listed in Table 8; |
2.Calculating disease kernel weights wd and gene kernel weights wg by Eq. 17 (CKA-MKL), respectively; |
3.Calculating \(\mathbf {K}^{*}_{d}\) and \(\mathbf {K}^{*}_{g}\) by Eq. 14, respectively; |
4.Calculating \(\mathbf {L}^{h}_{d}\) and \(\mathbf {L}^{h}_{g}\) by Eq. 21, respectively; |
5.Solving Eqs. 27 and 28 (ALSA), and estimating F∗ by Eq. 29; |