This is the third data set of submitted manuscript by Wu et al., which illustrates an assessment of 18 computational methods for inferring mutational signatures on 20 synthetic data sets that incorporated varying degrees of correlated activity of two common mutational signatures. Part III - Full results of 18 computational methods on 20 synthetic data sets when number of ground-truth signatures, 2, was not specified to the methods. This folder contains a .zip file for each approach tested with number of signatures, K, specified or initialized as 2. In addition to the zip files, this folder contains the file all_results_Kas2.csv, which contains an assessment of the results from every run of every approach on every data set. Broadly, each .zip file contains extracted signatures, inferred exposures, and other files generated from individual runs, as well as some analyses of the results. These are in folders named according to the scheme S.ratio.Rsq.correlation, where ratio is the SBS1-SBS5 exposure ratio, and correlation is the Pearson R2 for the correlation between SBS1 and SBS5 exposures. In more detail, each folder named S.ratio.Rsq.correlation contains the results of 20 runs for one spectra database (except in the folders for maftools and MutationalPatterns, which have 1 run for each dataset). See README.Wu.et.al.K.2.txt.