void GroupProteinsOrPeptides(
IDictionary <string, Node> entities,
IReadOnlyList <string> entityNames,
Type entityType,
GlobalIDContainer globalIDTracker)
{
for (var count = 0; count != entityNames.Count; count++)
{
// Is the key there?
if (!entities.ContainsKey(entityNames[count]))
{
continue;
}
// Get the protein or peptide
var entity = entities[entityNames[count]];
// Only proceed if the correct type
if (entity.GetType() != entityType)
{
continue;
}
var duplicates = new NodeChildren <Node>();
// Look for duplicates and add to a duplicate list
duplicates.AddRange(FindDuplicates(entity));
if (duplicates.Count <= 1)
{
continue;
}
// Create a protein or peptide group from the duplicates
Group newGroup;
if (entityType == typeof(Protein))
{
newGroup = new ProteinGroup(duplicates, globalIDTracker);
}
else if (entityType == typeof(Peptide))
{
newGroup = new PeptideGroup(duplicates, globalIDTracker);
}
else
{
throw new Exception("Invalid type: must be Protein or Peptide");
}
foreach (var duplicateItem in duplicates)
{
// Remove entities from the library, add the new group
entities.Remove(duplicateItem.NodeName);
}
entities.Add(newGroup.NodeName, newGroup);
}
}
private NodeChildren <Node> FindDuplicates(Node node)
{
node.Children.Sort();
var candidates = new NodeChildren <Node>();
candidates.AddRange(node.Children[0].Children);
candidates.Remove(node);
var count = 0;
// Pulls out candidates with different counts of children than the master
while (candidates.Count != count)
{
if (node.Children.Count != candidates[count].Children.Count)
{
candidates.RemoveAt(count);
}
else
{
count++;
}
}
// May want to change the List to a Hashset to be faster.
// Finds identical sets.
foreach (var childNode in node.Children)
{
count = 0;
while (candidates.Count != count)
{
if (!childNode.Children.Contains(candidates[count]))
{
candidates.RemoveAt(count);
}
else
{
count++;
}
}
}
candidates.Add(node);
return(candidates);
}
