Physical Chemistry Chemical Physics最新文献

We have investigated the dynamics of vibrational spectral diffusion, hydrogen bonds and orientational relaxation of water in glycerol-water mixtures of varying concentration. We have looked at how these water dynamical properties are affected by glycerol in different solvation environments through calculations of the linear and two dimensional infrared (2DIR) spectra, and various time correlation functions. We have focused on the low concentration regime with the glycerol mole fraction (x_GLC) going up to 0.12. It is found that the linear infrared spectra do not show any changes in the stretch frequencies of water in this concentration regime, since the OH groups of glycerol molecules provide a hydrogen bonding environment similar to that of water OH groups. However, the dynamics of spectral diffusion calculated from 2DIR spectra at the non-Condon level and the frequency time correlation function (FTCF) of water molecules that are present in the hydration shells of two or more glycerol molecules, referred to as the trapped water, show a noticeable change in the rate of slowing down beyond the glycerol mole fraction of x_GLC ≈ 0.075. A similar change in the dynamics is also observed for orientational relaxation of trapped water molecules at the same glycerol concentration, which we refer to as the cross-over concentration for these mixtures. The dynamics of bulk water and also of those in the hydration shell of a single glycerol molecule are found not to exhibit any such crossover with increase of glycerol concentration. The distinct dynamical behavior of trapped water with variation of glycerol concentration can be linked to the glycerol induced effects on the dynamics of water-water hydrogen bonds.

G-quadruplexes (G4) are non-canonical nucleic acid secondary structures formed in guanine-rich regions and have been shown to regulate diverse cellular processes such as gene expression, DNA replication, and telomere maintenance, with increasing evidence linking G4 to cancer and other human diseases. G4 predominantly emerge in guanine-rich regions and are implicated in a spectrum of molecular interactions and disease phenotypes, thus researchers are interested in the formation of G4. However, predicting the formation of G4 from nucleotide sequences is a persistent problem. Existing computational tools for G4 prediction are either rule-based on domain knowledge or rely on a single neural network model like a convolutional neural network (CNN), which lacks interpretability and struggles to capture long-range dependencies among bases. Here, we introduce TransG4, a novel neural network architecture that integrates a CNN, a transformer, and bidirectional gated recurrent units (BiGRUs) to identify potential G4 structures. TransG4 demonstrates strong predictive performance on both G4-seq and rG4-seq datasets, accurately predicting DNA mismatch scores and consistently outperforming existing methods in RNA RSR-ratio prediction. Attention-based interpretations further show that TransG4 captures biologically meaningful motifs consistent with canonical G4 structures, providing an interpretable and generalizable framework and representing a novel and impactful contribution to sequence-based G4 propensity prediction.

Congenital hypothyroidism can result from mutations in human iodotyrosine deiodinase (hIYD), which catalyzes the deiodination of iodotyrosines (I-Tyr), a key step in thyroid hormone synthesis. Three homozygous mutations (R101W, F105-I106L, and I116T) are known causes of hypothyroidism. This computational study reveals that of the two loop I mutations in the flavin-binding domain (R101W and F105-I106L), F105-I106L has a stronger effect, causing greater structural distortion and weaker packing at the dimerization interface. These mutations reduce the binding energy of flavin and I-Tyr, compared to the wild type, due to a complete loss of R101 crown-like phosphate hydrogen bond in R101W and a partial loss of R101 and R279 hydrogen bonds in F105-I106L. In contrast, the distal I116T mutation has a marginal structural effect, but it alters the solvent-accessible surface area, van der Waals packing, and side-chain flexibility, which may explain its delayed clinical onset. Although the I116T mutation is far from the active site, it strengthens flavin and substrate binding via long-range effects. Protein-folding analysis via the Wako-Saitô-Muñoz-Eaton model shows that the wt-hIYD and R101W fold through the C-terminal region, while F105-I106L and I116T alter the folding pathway. Mutation-specific disruptions can impair electron transfer by altering I-Tyr alignment and flavin ring planarity. These findings reveal how hIYD mutations cause structural, energetic, and catalytic defects linked to hypothyroidism.

The precise adjustment of pH in complex buffered systems represents a critical process in chemical synthesis and biopharmaceutical development. However, the intricate multi-buffer equilibria pose significant challenges for conventional methods, leading to modeling difficulties and low optimization efficiency. We have developed a low-cost automated titration platform (total hardware cost < 100 USD) and established a hybrid physics-informed active learning framework that achieves target pH values in as few as 3-5 experimental iterations. Validation across diverse buffer systems, including phosphate, acetate, citrate, and ammonium buffers, demonstrates substantial efficiency improvements over purely data-driven methods, with rapid convergence to the target pH achieved in minimal experimental iterations. Beyond its scientific contributions, this work also offers important pedagogical value by providing a low-cost, transparent, and modular platform that allows students and early-stage researchers to gain hands-on experience in automated experimentation, chemical equilibria, and machine learning.