Protein Groups Report
Creates a Protein Groups Report for a dataset or for a union of several sets. If sets are added to Right Datasets, additional comparisons are made between the left and the right dataset unions: lists of common and unique groups are constructed. Different grouping methods can be chosen. Cutoff parameters are for setting a threshold for proteins they must pass before taken into consideration when grouping. Non-passing proteins are treated as if they don't exist at all.
Grouping is made by using agglomerative hierarchical clustering. Similarity scores are calculated for each existing pair of groups. Best scoring group pair is combined to form a new group. Scores are calculated for the new group against all of the remaining old ones. Again, best scoring pair is chosen for combining. The process is repeated until there are no group pairs passing the thresholds expressed by setting the parameters.
By choosing the thresholds very loosely, it is possible to spend a great deal of time and computer memory when clustering. If you run out of memory or patience, try setting tighter parameters.
Tooltips: Hover your mouse over each parameter in the parameter dialog to get more detailed information.