Search for: All records

Thin metal particles with two-dimensional symmetry are attractive for multiple ap- plications, but are difficult to synthesize in a reproducible manner. Although molecules that selectively adsorb to facets have been used to control nanoparticle shape, there is still limited research into the temporal control of growth processes to control these structural outcomes. Moreover, much of the current research into the growth of thin two-dimensional particles lacks mechanistic details. In this work, we study why the substitution of isoleucine for methionine in a gold binding peptide (Z2, RMRMKMK) results in an increase in gold nanoparticle anisotropy. Nanoplatelet growth in the pres- ence of Z2M246I (RIRIKIK) is characterized using in situ small-angle X-ray scattering (SAXS) and UV-Vis spectroscopy. Fitting time-resolved SAXS profiles reveals that 10 nm thick particles with two-dimensional symmetry are formed within the first few min- utes of the reaction. Next, through a combination of electron diffraction and molecular dynamics simulations, we show that substitution of methionine for isoluecine increases the (111) facet selectivity in Z2M246I, and conclude that this is key to the growth of nanoplatelets. However, the potential application of nanoplatelets formed using Z2M246I is limited due to their uncontrolled lateral growth, aggregation, and rapid sedimentation. Therefore, we use a liquid handling robot to perform temporally con- trolled synthesis and dynamic intervention through the addition of Z2 to nanoplatelets growing in the presence of Z2M246I at different times. UV-Vis spectroscopy dynamic light scattering, and electron microscopy show that dynamic intervention results in control over the mean-size and stability of plate-like particles. Finally, we use in situ UV-Vis spectroscopy to study plate-like particle growth at different times of interven- tion. Our results demonstrate that both the selectivity and magnitude of binding free energy towards lattices is important for controlling nanoparticle growth pathways. 

Data science and machine learning are revolutionizing enzyme engineering; however, high-throughput simulations for screening large libraries of enzyme variants remain a challenge. Here, we present a novel but highly simple approach to comparing enzyme variants with fully atomistic classical molecular dynamics (MD) simulations on a tractable timescale. Our method greatly simplifies the problem by restricting sampling only to the reaction transition state, and we show that the resulting measurements of transition-state stability are well correlated with experimental activity measurements across two highly distinct enzymes, even for mutations with effects too small to resolve with free energy methods. This method will enable atomistic simulations to achieve sampling coverage for enzyme variant prescreening and machine learning model training on a scale that was previously not possible. 

Cellulases are largely afflicted by inhibition from their reaction products, especially at high-substrate loading, which represents a major challenge for biomass processing. This challenge was overcome for endoglucanase 1 (E1) from Acidothermus cellulolyticus by identifying a large conformational change involving distal residues upon binding cellobiose. Having introduced alanine substitutions at each of these residues, we identified several mutations that reduced cellobiose inhibition of E1, including W212A, W213A, Q247A, W249A and F250A. One of the mutations (W212A) resulted in a 47-fold decrease in binding affinity of cellobiose as well as a 5-fold increase in the kcat. The mutation further increased E1 activity on Avicel and dilute-acid treated corn stover and enhanced its productivity at high-substrate loadings. These findings were corroborated by funnel metadynamics, which showed that the W212A substitution led to reduced affinity for cellobiose in the +1 and +2 binding sites due to rearrangement of key cellobiose-binding residues.

 

Attention mechanisms have led to many breakthroughs in sequential data modeling but have yet to be incorporated into any generative algorithms for molecular design. Here we explore the impact of adding self-attention layers to generative β -VAE models and show that those with attention are able to learn a complex “molecular grammar” while improving performance on downstream tasks such as accurately sampling from the latent space (“model memory”) or exploring novel chemistries not present in the training data. There is a notable relationship between a model's architecture, the structure of its latent memory and its performance during inference. We demonstrate that there is an unavoidable tradeoff between model exploration and validity that is a function of the complexity of the latent memory. However, novel sampling schemes may be used that optimize this tradeoff. We anticipate that attention will play an important role in future molecular design algorithms that can make efficient use of the detailed molecular substructures learned by the transformer. 

Chemical engineering is being rapidly transformed by the tools of data science. On the horizon, artificial intelligence (AI) applications will impact a huge swath of our work, ranging from the discovery and design of new molecules to operations and manufacturing and many areas in between. Early adoption of data science, machine learning, and early examples of AI in chemical engineering has been rich with examples of molecular data science—the application tools for molecular discovery and property optimization at the atomic scale. We summarize key advances in this nascent subfield while introducing molecular data science for a broad chemical engineering readership. We introduce the field through the concept of a molecular data science life cycle and discuss relevant aspects of five distinct phases of this process: creation of curated data sets, molecular representations, data-driven property prediction, generation of new molecules, and feasibility and synthesizability considerations. 

Polylactic‐co‐glycolic acid (PLGA)‐basedpolymers are synthetic materials that are prominent in drug delivery. PLGA homopolymer is biodegradable, biocompatible and is often polymerized to polyethylene glycol (PEG) to form a block copolymer used to form core‐shell nanoparticles. PEG is known for reducing blood clearance and opsonization, in addition to imparting “stealth” properties to various drugs and biomaterials. Current formulation methodologies for PLGA–PEG copolymer nanoparticles can be tuned to control key parameters for improved therapeutic delivery; however, molecular‐level understanding of copolymer‐solvent interactions during nanoparticle formulation is lacking. Therefore, three different PLGA–PEG/solvent pairs are examined, in comparison to their homopolymer constituents, to better understand copolymerization effects and its impact on nanoparticle formulation. Results show that at room temperature PLGA–PEG oligomers in dimethyl sulfoxide are the most rigid in good solvent conditions (Flory exponent >0.5) and have the largest end‐to‐end relaxation times when compared to acetone and water. PEG has a Flory exponent of ~0.5 in both water and acetone, showing that the molecular dynamic model that is employed can reproduce its amphiphilic nature in solution. Knowledge of PLGA–PEG structure and dynamics can be used in the design of novel biomedical technologies that improve drug efficacy and reduce cost of treatment.