With 25 points in the cluster, sensitivity improved markedly to an overall rate of 75%, and 40 clustered points yielded an overall sensitivity of 91%.
For an overall comparison of genomes we selected the inflation value 3.1, because it gives a good compromise between clustering sensitivity and specificity.
The use of these methods instead of alignment methods or the classic known clustering methods by biologists can improve the clustering sensitivity and specificity and reduce significantly the computational time.
The toolkit allows many different definitions of the homology graph: it can be generated from arbitrary subsets of organisms in the database with arbitrary cutoffs and inflation parameters (clustering sensitivity), and three scoring metrics that emphasize different aspects of the protein pair homology (Additional file 3) [ 31, 39, 40].
The data involved in transactions can be classified into sensitive clusters against sensitivity and transaction types.
Note that for a hierarchical clustering the sensitivity of a clustering in level k is at least as large as the sensitivity of clusterings in higher levels while the specificity is at most as large as that of clusterings in higher levels.
Since clustering results are highly dependent on the initial number of clusters, the sensitivity of the AM was also examined as a function of the input number of clusters to find a 'suggestive' optimal number of clusters K* for the dataset [57].
When Gram-positive cocci were arranged in clusters, the sensitivity was 68.4%, specificity 97.8%, positive predictive value 88.1% and negative predictive value 92.8% for predicting the presence of Staphylococcus aureus in culture (p < 0.001).
The second measure we used is the geometric accuracy as introduced by Broh´ee and van Helden [ 22], which is the geometric mean of two other measures, namely the clustering-wise sensitivity (Sn) and the clustering-wise positive predictive value (PPV).
