Clustering Sequences

Process

1. Create a table and upload some fundamental data (140clusterlinkagecreate.pl)
2. Create the tables for storing clusters (200alignmentclustercreate.pl)
3. Cluster the sequence data (PartitionEngine.java or PartitionParameterEngine.java)
4. Review the clusters with visualization (DisplayEngine.java)

Remarks

The program PartitionEngine.java generates clusters for all sequences in one particular partition. Furthermore, several parameter are required.

1. PartitionParameterId - a partitionparameter_id of the partitionparameter table, which represents the result of one partitioning run
2. Partition - a partition of the partition table, which represents one partition
3. ScoringMatrixId - a clusterscoringmatrix_id of the the clusterscoringmatrix table, which identifies the scoring matrix to be used
4. LinkageId - a clusterlinkage_id of the clusterlinkage table, which identifies the linkage to be used for calculating consensus sequence scores
5. MinimumAlignmentLength
6. MinimumAlignmentScore

The program PartitionParameterEngine.java generates clusters for all sequences of one BLAST run. It requires the same parameters as PartitionEngine.java excluding the second one, which is Partition.

The program DisplayEngine.java generates a visualization for one PartitionEngine.java run. It requires an alignmentclusterhierarchyparameter_id of the alignmentclusterhierarchyparameter table as parameter, which represents the result of one PartitionEngine.java run.