Friday May 4th – Presentations

In by admin

Opening Remarks

David M. Rogers

University of South Florida

04:00 PM
Computational Chemistry

Using theory and experiment to elucidate the origin of product specificity in protein arginine methyltransferases

Orlando Acevedo1, Abhishek Thakur1, Tamar B. Caceres2, Owen M. Price2, Joan M. Hevel2

1 Department of Chemistry, University of Miami, Coral Gables, FL 33146.
2 Department of Chemistry and Biochemistry, Utah State University, Logan, UT 84322.

04:20 PM
Computational Chemistry

Protein arginine methyltransferases (PRMTs) play pivotal roles in signaling networks that control all aspects of central dogma, metabolism, and signal transduction. As the major arginine methylation enzyme, protein arginine methyltransferase 1 (PRMT1) strictly generates monomethylated arginine (MMA) and asymmetric dimethylated arginine (ADMA), but not symmetric dimethylated arginine (SDMA). However, it remains unclear how PRMT1 product specificity is regulated, an important exploit since PRMTs have been implicated in many human diseases. In our joint theoretical and experimental study, it was discovered for the first time that a single amino acid mutation (Met48 to Phe) in the PRMT1 active site enabled PRMT1 to generate MMA, ADMA, and a limited amount of SDMA. A double mutant H293S-M48F-PRMT1 produced SMDA as the major product with limited amounts of MMA and ADMA. Intriguingly, protein arginine methyltransferase 7 (PRMT7) is unique within the PRMT family in that is capable of only forming MMA. Our recent efforts determined that a mutation of Phe71 to Ile in PRMT7 allowed a second turnover to occur, similar to PRMT1. This seminar will highlight our efforts integrating biochemical, biophysical, computational, and structural approaches to provide mechanistic understanding of how the PRMT isoforms establish the strict product specificity that is required for the desired biological effect.

Atomistic insight towards fragmentary interactions of PEG in bioconjugates – an atomistic molecular dynamics study

Aravinda Munasinghe1, Akash Mathavan1,2, Akshay Mathavan1,2, Ping Lin1 and Coray M. Colina1,2

1 Department of Chemistry, University of Florida, Gainesville, FL 32611
2 Department of Materials Science and Engineering, University of Florida, Gainesville, FL 32611

04:50 PM
Computational Chemistry

Currently many polymers and their different architectures are being investigated to synthesize bioconjugates. Due to its biocompatibility, polyethylene glycol (PEG) has emerged as an interesting polymer for bio-conjugate synthesis. Though PEG has been extensively studied, how its interact with biomolecules is yet to be fully clarified. The purpose of this work was to explore how PEG interacts with the Bovine Serum Albumin (BSA) protein using atomistic molecular dynamics simulations. To understand conditions which promote PEG – BSA interactions, conjugated systems with different molecular weights (i.e., 2, 5, 10, and 20 kDa) and unbound PEG were studied. PEGylated polymers were conjugated to N-terminal as well as Lys116 to explore the effect of the conjugate site. In each system, contacts between the polymer and the protein were monitored to explore PEG – BSA interactions and the affinity of PEG towards the BSA was evaluated based on the total number of formed contacts. It was found that the affinity of PEG towards the protein surface increases as a function of molecular weight. Further analysis of these interactions has revealed that PEG could adopt extended or coil-like conformations near the protein surface where mainly hydrophobic or hydrophilic residues are predominant.

Free Energy Sampling of Long-Timescale Biomolecular Dynamics: The Orthogonal Space Sampling Paradigm

Lianqing Zheng, Dongsheng Wu, Karen Corbett, Erick Aitchison, Steven Austin, Xubin Li, Chao Lv, William Harris, and Wei Yang

Department of Chemistry and Biochemistry & Institute of Molecular Biophysics
Florida State University

05:45 PM
Computational Chemistry

In the past decades, various enhanced sampling schemes and methods have been proposed towards the dream of practical sampling of free energy surfaces underlying biologically relevant biological processes. Despite victories declared by various mathematical methods, their ineffectiveness has been increasingly clear; physical basis of their limitation, particularly from the viewpoint of energy flow acceleration, is also more obvious to the community. At the same time, a physics-centered scheme, the orthogonal space sampling (OSS) theory, has been emerging. The recent development and large-scale tests shows that robust free energy sampling of long-timescale biomolecular processes has been achieved. This talk will serve as a part of official announcement of this breakthrough.

Machine learning approaches to evaluate correlation patterns in allosteric signaling: a case study of the PDZ2 domain

Mohsen Botlani, Ahnaf Siddiqui, Sameer Varma

University of South Florida

06:10 PM
Computational Chemistry

Many proteins are regulated by dynamic allostery wherein regulator-induced changes in structure are comparable with thermal fluctuations. Consequently, understanding their mechanisms requires assessment of relationships between and within conformational ensembles of different states. Here we show how machine learning based approaches can be used to simplify this high-dimensional data mining task and also obtain mechanistic insight. In particular, we use these approaches to investigate two fundamental questions in dynamic allostery. Firstly, how do regulators modify inter-site correlations in conformational fluctuations ($C_{ij}$)? Secondly, how are regulator-induced shifts in conformational ensembles at two different sites in a protein related to each other? We address these questions in the context of the human PTP1E's PDZ2 domain, which is a model protein for studying dynamic allostery. We use molecular dynamics to generate conformational ensembles of the PDZ2 domain in both the regulator-bound and regulator-free states. The employed protocol reproduces methyl deuterium order parameters from NMR. Results from unsupervised clustering of $C_{ij}$ combined with flow analyses of weighted graphs of $C_{ij}$ show that regulator binding significantly alters the global signaling network in the protein; however, not by altering the spatial arrangement of strongly interacting amino acid clusters, but by modifying the connectivity between clusters. Additionally, we find that regulator-induced shifts in conformational ensembles, which we evaluate by repartitioning ensembles using supervised learning, are, in fact, correlated. This correlation $\Delta_{ij}$ is less extensive compared to $C_{ij}$, but in contrast to $C_{ij}$, $\Delta_{ij}$ depends inversely on the distance from the regulator binding site. Assuming that $\Delta_{ij}$ is an indicator of the transduction of the regulatory signal leads to the conclusion that the regulatory signal weakens with distance from the regulatory site. Overall, this work provides new approaches to analyze high-dimensional molecular simulation data, and also presents applications that yield new insight into dynamic allostery.

ANI strikes again. New results from a grown-up Machine learning method for organic systems.

Adrian E. Roitberg

University of Florida

06:35 PM
Computational Chemistry

In the theoretical study of molecular systems, a compromise between speed and accuracy is required to study the energetics of chemical systems. Quantum mechanical (QM) methods allow accurate energies and forces to be calculated but require substantial computational effort. Classical force fields are fast but only accurate near equilibrium and are generally unable to be used in reactivity studies due to their restrictive functional form.

Machine learning methods such as artificial neural networks have been used to develop neural network potentials (NNP), which are fitted to QM reference energies, though few have shown to be size extensible. Through the continued development of our methodology and data set, known as ANAKIN-ME (or ANI for short), we developed a new class of NNP, which are size extensible and chemically accurate. Specifically, we develop the ANI-1 potential for organic molecules containing H, C, N, and O. Through extensive benchmarks, case studies, and molecular dynamics simulations, we will provide evidence that the ANI method produces chemically accurate and size extensible potentials.

We will also show that active learning techniques allows these networks to learn new chemistry with very small amounts of new data, and that chemical reactions can then be studied with high accuracy.

As the results clearly show, these types of methods are a potential game changer for molecular simulations. The ANI method continues to bring a new, highly efficient, and accurate method for the development of NNPs into the realm of reality, and opens the door for the next generation of “out-of-the-box” general purpose potentials.