DumpHash.c geneid v 1.1 source documentation


Description:
Between one fragment of DNA input sequence and the next one to be processed, some useful information must be saved to go the prediction on: best partial genes and exons still not used in the gene assembling. But between best partial genes, there is a lot of redundancy (the same or similar set of exons). Every exon must be saved only once, not more, so that by using the dumspter (hash table of backup exons), is very easy to know whether one exon has already been copied or not. Exon features as signal positions, exon type or strand are used to bind unique keys to these exons.
Briefing:
void resetDumpHash(dumpHash* h)
Initialize the hash table and set the counter of exons copied.
long fDump(exonGFF* E)
Hash function: integer computed from exon features.
void setExonDumpHash(exonGFF* E, dumpHash* h)
Insert the input exon into the dumpster hash table after have been copied.
exonGFF* getExonDumpHash(exonGFF* E, dumpHash* h)
Finding an exon into the hash table to know it must be whether copied or not. If it has already been copied before, the address of the copy is returned.
void freeDumpNodes(dumpNode* node)
Free recursively a list of sinonimous nodes.
void cleanDumpHash(dumpHash *h)
Free all of lists of sinonimous nodes in the hash table.




Enrique Blanco Garcia © 2001