Skip to main content
Figure 2 | BMC Genomics

Figure 2

From: PNAC: a protein nucleolar association classifier

Figure 2

Generation of the training and testing sets. Two datasets were used to generate the training and testing sets. A manually curated literature-based nucleolar association dataset (blue list) was used to construct the training set (which is also used in the leave-one-out cross validation test and referred to as the test set 1) and a non-overlapping independent literature-based test set (test set 2). An experimental SILAC dataset (red list) was used to construct the independent SILAC-derived test set (test set 3). The intersection of the manually curated literature dataset (blue list) and the experimental SILAC dataset (red list) is shown in purple and was used to map the SILAC data points to our nucleolar association groups to create the SILAC test set. The generation of the training and testing sets is described in more detail in the Methods section.

Back to article page