CAREER: Enhanced Sampling Methods to Characterize Nucleic Acid Structure, Recognition Mechanisms and Function

点击翻译按钮获取中文摘要

With support from the Chemical Theory, Models and Computational Methods (CTMC) program in the Division of Chemistry, Alberto Perez of the University of Florida will develop new computational methods directed at better understanding nucleic acids. Nucleic acids are the repository of genetic information and have numerous functional roles in biology as well as having applications in fields such as nanotechnology, nanoelectronics and sensor technology. Despite their importance, the tool kit is still limited to enable the successful modeling and prediction of the most likely conformations (three dimensional disposition of atoms) nucleic acids adopt. Alberto Perez and his research group will pursue the development of new computational tools that address challenges in the structure prediction of DNA (2'-deoxyribonucleic acid), RNA (ribonucleic acid), and their complexes with proteins. The new methods are designed to connect the structures of these molecules with their functional roles to help decipher their mechanisms of action. The tools are to be made freely available to the community to increase their impact on chemical biology and biotechnology. Additionally, Dr. Perez and his group will train new generations of scientists through immersive virtual reality activities that place students in the role of molecular modelers. These interactive modules will enable students to develop STEM-related skills through an inquiry-based approach. The highly charged and flexible nature of nucleic acids has hindered progress in the development of docking and simulation tools to sample conformational changes, including those involved in three dimensional folding and binding. Alberto Perez and his team aim to address the problem of sampling large conformational transitions in nucleic acids and to derive a physico-chemical understanding of sequence-function relationships. To achieve this goals, Dr. Perez and his group will develop approaches based on generalized ensemble methods that sample large conformational changes, especially those involving a phase transition (e.g., binding or folding). They will further develop new strategies based on Bayesian inference to leverage experimental data that is insufficient on its own to determine the structures of nucleic acids and their complexes. These methodologies are expected increase sampling efficiency, robustness, and reproducibility of simulation results. Finally, to provide guidance for the combinatorial explosion of available nucleic acid sequences and derive a physico-chemical understanding of structure-function relationships that scale with sequence length, Dr. Perez and his team will develop and apply methods based on Markov State and Dynamical Graphical Models. If successful, in the longer term, these methods will help rationalize the mechanism by which protein-nucleic acid recognition takes place.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.