Common use of Removing redundancy Clause in Contracts

Removing redundancy. ‌ In order to compute the CAT curves, and then compare the results obtained from the three experiments, it is necessary to identify the set of non-redundant features in common among the three data sets. The filterRedundant function enables to remove the redundant features within each data.frame prior computing the subset of of common identifiers across the three data sets. The argument idCol specifies the index or the name of the column containing redundant identifiers, while the other arguments enable to control the method used to remove the redundancy, as explained in the help for this function. By default the filterRedundant function will keep the feature with the largest absolute ranking statistics, as defined by the byCol argument. The method argument enables to con- trol which feature to keep and which to discard. Currently available methods are: maxORmin, geoMean, random, mean, and median (see help for filterRedundant). In the example below we are using an lapply call to remove the redundant features from all the three data.frames at once. To this end we will use gene symbols to identify the redundant features and the t-statistics to select which feature to keep. Each data.frame now contains only unique features. dataSetA dataSetB dataSetC [1,] 1075 2058 1500

Removing redundancy. ‌ In order to compute the CAT curves, and then compare the results obtained from the three experimentsexper- iments, it is necessary to identify the set of non-redundant features in common among the three data sets. The filterRedundant function enables to remove the redundant features within each data.frame prior computing the subset of of common identifiers across the three data sets. The argument idCol specifies the index or the name of the column containing redundant identifiers, while the other arguments enable to control the method used to remove the redundancy, as explained in the help for this function. By default the filterRedundant function will keep the feature with the largest absolute ranking rank- ing statistics, as defined by the byCol argument. The method argument enables to con- trol control which feature to keep and which to discard. Currently available methods are: maxORmin, geoMean, random, mean, and median (see help for filterRedundant). In the example below we are using an lapply call to remove the redundant features from all the three data.frames at once. To this end we will use gene symbols to identify the redundant features and the t-statistics to select which feature to keep. Each data.frame now contains only unique features. dataSetA dataSetB dataSetC [1,] 1075 2058 1500