Applying machine learning techniques to DNA sequence analysis. Progress report, Year 2, February 14, 1992--December 11, 1992

摘要

We are primarily developing a machine teaming (ML) system that modifies existing knowledge about specific types of biological sequences. It does this by considering sample members and nonmembers of the sequence motif being teamed. Using this information, our teaming algorithm produces a more accurate representation of the knowledge needed to categorize future sequences. Specifically, our KBANN algorithm maps inference rules about a given recognition task into a neural network. Neural network training techniques then use the training examples to refine these inference rules. We call these rules a domain theory, following the convention in the machine teaming community. We have been applying this approach to several problems in DNA sequence analysis. In addition, we have been extending the capabilities of our teaming system along several dimensions. We have also been investigating parallel algorithms that perform sequence alignments in the presence of frameshift errors.