Gene Estimation List constructing algorithm is employed for generating an estimation list for each gene with missing values, consisting of genes with expression profiles which exhibit at least minimum relative (preliminary defined) similarity in terms of DTW distance to a target expression profile (one of the gene with missing values). These profiles are consequently used for missing values estimation.

In order to estimate missing values in any location of gene g_i, a set of genes with complete expression profiles (no missing values) at a maximum R-relative radius DTW distance (R is preliminary defined) from the profile of g_i needs to be identified. In this process, all genes with non-missing values are considered and a gene estimation list, which is further used to impute the missing values of gene g_i, is constructed.

Gene Estimation List

The values of the genes in the gene estimation list can be used in various fashions, see DTWimpute algorithm, to estimate missing values of gene g_i. The contribution of each gene in the estimation list is weighted by the degree of similarity (based on its DTW distance) of its expression profile to that of gene g_i.