14 October 2014 in Proceedings of the National Academy of Sciences
Significance: Understanding the interdependency of physiochemical properties of nanomedicine (NM) in correlation to its biological response and function is crucial for additional development of anticancer NM. Here, we prepared monodisperse drug–silica nanoconjugates in three distinct sizes (20, 50, and 200 nm) with other physiochemical properties controlled to be identical to investigate size-dependent biodistribution, tumor tissue penetration and clearance, and anticancer efficacy in various tumor models. We also developed a mathematical model of the spatiotemporal distribution of NM within a tumor to gain insight into the size-dependent interaction with tumor. Our studies show clear evidence that there is an optimal size of anticancer NM and that NM with the optimal size has the highest tumor retention integrated over time.
30 September 2018 in Journal of Computational Chemistry
16 July 2020 in Journal of Chemical Theory and Computation
12 October 2015 in Proceedings of the National Academy of Sciences
Significance: We developed antimicrobial polypeptides (AMPs) with unprecedented radial amphiphilicity. Unlike typical AMPs characterized by facial amphiphilicity or biomimetic antimicrobial polymers with randomly distributed charged and hydrophobic groups, this class of AMPs is made up of homo-polypeptides that feature a radially amphiphilic (RA) structure and adopt a stable α-helical conformation with a hydrophobic helical core and a charged exterior shell, which is formed by flexible hydrophobic side chains with terminal charge group. The RA polypeptides appear to offer several advantages over conventional AMPs with regard to stability against proteases and simplicity of design. They also exhibit high antibacterial activity against both gram-negative and gram-positive bacteria and low hemolytic activity. This design may become a general platform for developing AMPs to treat drug-resistant bacteria.
7 June 2019 in The Journal of Chemical Physics
The success of enhanced sampling molecular simulations that accelerate along collective variables (CVs) is predicated on the availability of variables coincident with the slow collective motions governing the long-time conformational dynamics of a system. It is challenging to intuit these slow CVs for all but the simplest molecular systems, and their data-driven discovery directly from molecular simulation trajectories has been a central focus of the molecular simulation community to both unveil the important physical mechanisms and drive enhanced sampling. In this work, we introduce state-free reversible VAMPnets (SRV) as a deep learning architecture that learns nonlinear CV approximants to the leading slow eigenfunctions of the spectral decomposition of the transfer operator that evolves equilibrium-scaled probability distributions through time. Orthogonality of the learned CVs is naturally imposed within network training without added regularization. The CVs are inherently explicit and differentiable functions of the input coordinates making them well-suited to use in enhanced sampling calculations. We demonstrate the utility of SRVs in capturing parsimonious nonlinear representations of complex system dynamics in applications to 1D and 2D toy systems where the true eigenfunctions are exactly calculable and to molecular dynamics simulations of alanine dipeptide and the WW domain protein.