public class TagMatchFinder
Very simple but fast homology finder. Very similar to BLAT, with long word searches. A in memory index is created from any set of Tags, and this list many then be quickly queried for high homology hits.
With the current code no indels will be discovered, however, this could be easily changed. Currently, tuned to a word length of 16. If you are going to use this please read this code.
The lookup match tables are based on 2-bit encoded long sequences, and the query also needs to in a 2-bit long
public static int wordLength
public static int maxDivergence
public TagMatchFinder(Tags theTags)
public kotlin.Array[] getTagLookTable()
public static void main(java.lang.String[] args)
public java.util.TreeMap<java.lang.Integer,java.lang.Integer> findMatchesWithIntLengthWords(kotlin.Array[] query,
int maxDiv,
boolean keepOnlyBest)
Return a TreeMap good hits based on a sequence query. The returned tree map is the list of tag indices as key, divergence as value. Ed- It seems like the key & value should perhaps be reversed.
query - array of 2-bit encoded long querymaxDiv - maximum divergence to look forkeepOnlyBest - result only includes the best result