Brooke Lustig

RNA base-amino acid interaction strengths derived from structures and sequences.

We investigate RNA base-amino acid interactions by counting their contacts in structures and their implicit contacts in various functional sequences where the structures can be assumed to be preserved. These frequencies are cast into equations to extract relative interaction energetics. Previously we used this approach in considering the major groove interactions of DNA, and here we apply it to the more diverse interactions observed in RNA. Structures considered are the three different tRNA synthetase complexes, the U1A spliceosomal protein with an RNA hairpin and the BIV TAR-Tat complex. We use binding data for the base frequencies for the seryl, aspartyl and glutaminyl tRNA-synthetase and U1 RNA-protein complexes. We compare with the previously reported DNA major groove peptide contacts the results for atoms of RNA bases, usually in the major groove. There are strong similarities between the rank orders of interacting bases in the DNA and the RNA cases. The apparent strongest RNA interaction observed is between arginine and guanine which was also one of the strongest DNA interactions. The similar data for base atomic interactions, whether base paired or not, support the importance of strong atomic interactions over local structure considerations, such as groove width and alpha-helicity.

Flexibility of BIV TAR-Tat: Models of peptide binding

Abstract A new approach in determining local residue flexibility from base-amino acid contact frequencies is applied to the twelve million lattice chains modeling BIV Tat peptide binding to TAR RNA fragment. Many of the resulting key features in flexibility correspond to RMSD calculations derived from a set of five NMR derived structures (X. Ye, R. A. Kumar, and D. J. Patel, Protein Data Bank: Database of three-dimensional structures determined from NMR (1996)) and binding studies of mutants (L. Chen and A. D. Frankel, Proc. Natl. Acad. Sci. USA 92, 5077–5081 (1995)). The lattice and RMSD calculations facilitate the identification of peptide hinge regions that can best utilize the introduction of Gly or other flexible residues. This approach for identifying potential sites amenable to substitution of more flexible residues to enhance peptide binding to RNA targets could be a useful design tool.

Biological applications of hammerhead ribozymes as anti-viral molecules

Ribozymes are catalytic RNAs that can cleave substrate RNAs in a sequence specific manner. Here we survey, in brief, the structure of hammerhead and hairpin ribozymes and discuss their applications as molecular antiviral molecules for HIV-1.

Logistic regression models to predict solvent accessible residues using sequence- and homology-based qualitative and quantitative descriptors applied to a domain-complete X-ray structure learning set.

This article describes the development, validation and application of simple logistic regression models for the prediction of solvent accessible residues in oligomer and non-oligomer sets using a domain-complete X-ray structure learning set.

Novel application of query-based qualitative predictors for characterization of solvent accessible residues in conjunction with protein sequence homology. Proceedings of the 22nd International Workshop on Database and Expert Systems Applications

Prediction of relative solvent accessibility (RSA) is a standard first-approach in predicting threedimensional protein structures. Here we have applied linear regression methods that include various sequence homology values for each residue as well as query residue qualitative predictors, corresponding to each of the twenty canonical amino acids. We fit the 268-protein learning set with a variety of sequence homology terms, including 20 and 6-term sequence entropy, and residue qualitative predictors. Then estimated RSA values are subsequently generated for the 215-protein Manesh test set. The qualitative predictors describe the actual query residue type (e.g. Gly) as opposed to the measures of sequence homology for the aligned subject sequences. This is consistent with our framework of modeling a limited set of discrete and/or physically intuitive predictors. Initial calculations involving normalized RSA values were considered as a likely first attempt, incorporating the notion of fitting an explicit binary characterization of individual residues, either as buried or accessible. Interestingly, the utilization of qualitative predictors showed significant prediction accuracy. Subsequent calculations using the original RSA values gave estimated values that, upon binary classification, indicated accuracies comparable to other first stage methods. Development of a second stage methodology is of current interest. Keywords-hydrophobicity, sequence entropy, buried residues, surface accessibilities, qualitative predictors

Novel Application of Query-Based Qualitative Predictors for Characterization of Solvent Accessible Residues in Conjunction with Protein Sequence Homology

Prediction of relative solvent accessibility (RSA) is a standard first-approach in predicting three-dimensional protein structures. Here we have applied linear regression methods that include various sequence homology values for each residue as well as query residue qualitative predictors, corresponding to each of the twenty canonical amino acids. We fit the 268-protein learning set with a variety of sequence homology terms, including 20 and 6-term sequence entropy, and residue qualitative predictors. Then estimated RSA values are subsequently generated for the 215-protein Manesh test set. The qualitative predictors describe the actual query residue type (e.g. Gly) as opposed to the measures of sequence homology for the aligned subject sequences. This is consistent with our framework of modeling a limited set of discrete and/or physically intuitive predictors. Initial calculations involving normalized RSA values were considered as a likely first attempt, incorporating the notion of fitting an explicit binary characterization of individual residues, either as buried or accessible. Interestingly, the utilization of qualitative predictors showed significant prediction accuracy. Subsequent calculations using the original RSA values gave estimated values that, upon binary classification, indicated accuracies comparable to other first stage methods. Development of a second stage methodology is of current interest.

The G2 Project: Establishing A CSU “Grand Grid” for Scientific Computing in Research and Education

The G2 project reflects a systematic effort to functionally link computational resources across all California Sate University (CSU) campuses. The result is a CSU wide Grand Grid that provides a computing ecosystem where Computational Resources within and outside CSU can join in fully or partially while they maintain their autonomy. The Grand Grid provides interfaces for secure and efficient interaction among distinct administrative domains depending on the classification of the user accessing a service on the Grid and the hosting site(s) of the applications supporting the service. The current computational state, the architectural principles, and a high level blueprint of the Grand Grid are presented in this paper. Globus Toolkit v4 is presented as the leading candidate for middleware on which the Grand Grid is being built, and an initial set of participating CSU computational resources that help in shaping the Grand Grid is introduced.