Matrix imputation in gene expression data using DCT based sparsity
Matlab codes for matrix imputation-
- DCT_Stage1.m - Code for imputing missing values using only Stage-1 i.e DCT based sparsity
- CTNNBI.m - Code for introducing missing values randomly and then imputing missing values using DCT sparsity in Stage-1 and Nuclear Norm minimization based denoising in Stage-2 and finally calculating NRMSE
- TV_DCTcode2.m - Code for introducing missing values randomly and then imputing missing values using DCT based sparsity in Stage-1 and TV denosing in Stage-2 and finally calculating NRMSE
- TV_DCTcode1.m - Code for imputing missing values in incomplete matrix using TV-DCT algorithm
%% NOTE: please install spgl solver and keep it in your path in matlab for this algorithm to work
Please cite our work if you use our codes for matrix imputation For TV-DCT method cite the following- -> Farswan, A. and Gupta, A., 2019, May. TV-DCT: Method to Impute Gene Expression Data Using DCT Based Sparsity and Total Variation Denoising. In ICASSP 2019-2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (pp. 1244-1248). IEEE.
For CT-NNBI method, cite the following- -> Gehlot S, Farswan A, Gupta A, Gupta R. CT-NNBI: Method to Impute Gene Expression Data using DCT Based Sparsity and Nuclear Norm Constraint with Split Bregman Iteration. In2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019) 2019 Apr 8 (pp. 1315-1318). IEEE.