All of the projects in this proposal involved studies of specificity of protein/nucleic acid interactions. We want to know what sequence patterns are recognized by several different proteins (or complexes) and quantitatively how the recognition changes with changes in sequence. The ribosome binding site project continues our work at understanding how E. coli ribosomes interact with initiation sites. We will determine the quantitative contributions of ribosome binding site sequences features, such as the Shine/Dalgarno sequence, the initiation codon and other important bases, and the quantitative effects of mRNA structure on initiation rates. The DNA binding studies will use several proteins, T7 RNA polymerase and cI, cro and mnt regressors to examine quantitative sequence specificity. These studies will use both established methods for determining sequence specificities and a new method for rapidly determining the changes in binding energy associated with changes in binding site sequence. This method follows from our investigation of information theory applications to sequence analysis. Finally, we will continue our work in developing computer tools for sequence analysis and specifically for the generation of specificity representations of proteins/nucleic acid interactions. We will develop a new method for finding the binding pattern of a protein from a collection of unaligned sequences, the """"""""consensus problem""""""""/ that is both powerful and efficient.
Showing the most recent 10 out of 36 publications
