Performance Evaluation of L1-norm-based Microarray Missing Value Imputation

l1-norm minimization was utilized in the imputation of microarray missing values, which is an important procedure in bioinformatics experiments. Two l1 approaches, based on the framework of local least squares (LLS) and iterative biclusterbased least squares (bicluster-iLLS) respectively, were employed. Imputed datasets of the l1 approaches were compared with those of traditional l2 methods. The imputation error rates showed that the assumption of sparsity is not supported by the microarray datasets. Singular value decompositions in biclusters and in the neighborhoods of target genes were computed to show the structure of a microarray dataset. The coefficients of l1 minimization solutions were also analyzed to reveal possible reasons for the performance of l1 approaches. Keywords-microarray missing value imputation; sparse representation; l1 minimization