Ligand Characterization and DNA Intercalation of Ru(II) Polypyridyl Complexes: A Local Vibrational Mode Study.

We investigated in this work ruthenium-ligand bonding across the RuN framework in 12 Ru(II) polypyridyl complexes in the gas phase and solution for both singlet and triplet states, in addition to their affinity for DNA binding through π-π stacking interactions with DNA nucleobases. As a tool to assess the intrinsic strength of the ruthenium-ligand bonds, we determined local vibrational force constants via our local vibrational mode analysis software. We introduced a novel local force constant that directly accounts for the intrinsic strength of the π-π stacking interaction between DNA and the intercalated Ru(II) complex. According to our findings, [Ru(phen)2(dppz)]2+ and [Ru(phen)2(11-CN-dppz)]2+ provide an intriguing trade-off between photoinduced complex excitation and the strength of the subsequent π-π stacking interaction with DNA. [Ru(phen)2(dppz)]2+ displays a small singlet-triplet splitting and a strong π-π stacking interaction in its singlet state, suggesting a favorable photoexcitation but potentially weaker interaction with DNA in the excited state. Conversely, [Ru(phen)2(11-CN-dppz)]2+ exhibits a larger singlet-triplet splitting and a stronger π-π stacking interaction with DNA in its triplet state, indicating a less favorable photoinduced transition but a stronger interaction with DNA postexcitation. We hope our study will inspire future experimental and computational work aimed at the design of novel Ru-polypyridyl drug candidates and that our new quantitative measure of π-π stacking interactions in DNA will find a general application in the field.

LAST: Latent Space-Assisted Adaptive Sampling for Protein Trajectories.

Molecular dynamics (MD) simulation is widely used to study protein conformations and dynamics. However, conventional simulation suffers from being trapped in some local energy minima that are hard to escape. Thus, most of the computational time is spent sampling in the already visited regions. This leads to an inefficient sampling process and further hinders the exploration of protein movements in affordable simulation time. The advancement of deep learning provides new opportunities for protein sampling. Variational autoencoders are a class of deep learning models to learn a low-dimensional representation (referred to as the latent space) that can capture the key features of the input data. Based on this characteristic, we proposed a new adaptive sampling method, latent space-assisted adaptive sampling for protein trajectories (LAST), to accelerate the exploration of protein conformational space. This method comprises cycles of (i) variational autoencoder training, (ii) seed structure selection on the latent space, and (iii) conformational sampling through additional MD simulations. The proposed approach is validated through the sampling of four structures of two protein systems: two metastable states of Escherichia coli adenosine kinase (ADK) and two native states of Vivid (VVD). In all four conformations, seed structures were shown to lie on the boundary of conformation distributions. Moreover, large conformational changes were observed in a shorter simulation time when compared with structural dissimilarity sampling (SDS) and conventional MD (cMD) simulations in both systems. In metastable ADK simulations, LAST explored two transition paths toward two stable states, while SDS explored only one and cMD neither. In VVD light state simulations, LAST was three times faster than cMD simulation with a similar conformational space. Overall, LAST is comparable to SDS and is a promising tool in adaptive sampling. The LAST method is publicly available at https://github.com/smu-tao-group/LAST to facilitate related research.

The Local Vibrational Mode Theory and Its Place in the Vibrational Spectroscopy Arena.

This Feature Article starts highlighting some recent experimental and theoretical advances in the field of IR and Raman spectroscopy, giving a taste of the breadth and dynamics of this striving field. The local mode theory is then reviewed, showing how local vibrational modes are derived from fundamental normal modes. New features are introduced that add to current theoretical efforts: (i) a unique measure of bond strength based on local mode force constants ranging from bonding in single molecules in different environments to bonding in periodic systems and crystals and (ii) a new way to interpret vibrational spectra by pinpointing and probing interactions between particular bond stretching contributions to the normal modes. All of this represents a means to work around the very nature of normal modes, namely that the vibrational motions in polyatomic molecules are delocalized. Three current focus points of the local mode analysis are reported, demonstrating how the local mode analysis extracts important information hidden in vibrational spectroscopy data supporting current experiments: (i) metal-ligand bonding in heme proteins, such as myoglobin and neuroglobin; (ii) disentanglement of DNA normal modes; and (iii) hydrogen bonding in water clusters and ice. Finally, the use of the local mode analysis by other research groups is summarized. Our vision is that in the future local mode analysis will be routinely applied by the community and that this Feature Article serves as an incubator for future collaborations between experiment and theory.