engleski

A reduced gene database for precision species detection

For the purpose of a faster identification of sequences, the current database should be modeled in a different way. The current databases should be converted to the reduced database. The reduced database is modeled in a way that at first it gives us information only about significant sequences, and all the other sequences are hidden behind their representative. We proposed three different approaches for creating reduced database. Two of them incorporates hierachical clustering, while the third one is based on partitional clustering. Since the approach with partitional clustering has been shown as the best, our final design contains implementation of it. Two programs has been created, the Reducer and the Searcher. The Reducer creates a reduced database, while the Serchers search through the reduced database to find a new sequence.

Clustering; identification; Reduced database

nije evidentirano