Training in Sequence Analysis Using Supercomputers

Carnegie-Mellon University, Pittsburgh PA

Investigators

Linked publications, trials & patents

Abstract

[unreadable] DESCRIPTION: (provided by applicant) The principal aim of the course is to significantly enhance the productivity of experimental molecular biologists by training them in the computational techniques necessary to extract the maximum amount of information from macrornolecular sequences. Such sequences are a major source of data in molecular biology and genomics and are available to experimentalists from their own experimental activities, from sequence databases, structure databases, and repositories for the complete genomes of individual species The core techniques in which we will train experimentalists are collectively known as multiple sequence analysis and are a particularly powerful adjunct to experimental technique of site-directed mutagenesis. We supplement these core multiple sequence analysis techniques with techniques for integrating the results of these analyses with known structures of proteins and nucleic acids. [unreadable] [unreadable] The computational techniques are organized around methods for both global and local multiple sequence alignment and for identifying informative patterns in families of sequences, either based on these alignments or identified by analysis of unaligned sequences. Recent presentations of the weeklong course emphasized advanced techniques for discovering informative patterns in groups of sequences that have not been aligned and which may be unaligned. Also stressed were how to use such patterns to identify additional macromolecular sequences related to those under investigation, relating those patterns to structural motifs and their functional correlates and how the information discovered can be used to guide laboratory experiments. Future editions of the course will also include information on the computational complexities of analyzing complete genomes and microarray data.