20 August 2025 in Nature Chemistry
Among all enzymatic metallocofactors, those found in nitrogenases, the P and L or M clusters, stand out for their intricate structures. They are assembled by proteins of the Nif gene cluster from Fe2S2 rhombs—the smallest building blocks in FeS cluster chemistry—through a sequence of reactions constructing a Fe8S8 precursor. To advance our understanding of how enzymes selectively build such elaborate inorganic molecules, here we parallel the biosynthetic pathway by reporting the rational stepwise assembly of [Fe8S8]
m+
(m = 2, 4, 6) clusters from [Fe2S2]2+ rhombs within an extensive cyclic synthetic network. A [Fe8S8]4+ cluster of unique topology is identified, for which we coin the term ‘interlocked’ double cubane. As a molecular analogue of the NifB K cluster, a proposed precursor to both the P and L or M clusters, its preparation and the characterization of all related intermediates, offers fundamental insights into the molecular mechanisms governing the assembly of both biogenic and synthetic FeS clusters.
2 February 2026 in ChemRxiv
The consideration of transition state (TS) conformer ensembles is required to accurately model a reaction, and thus plays a key role in computational catalyst design. While CREST and GOAT are established methods for TS conformer ensemble generation, the associated computational cost remains a major bottleneck in computational chemistry pipelines, including for the generation of large machine learning datasets for catalyst design. To this end, we present racerTS ( RA pid C onformer E nsembles with R dkit for T ransition S tates), a method for efficient TS conformer ensemble generation. In this work, we describe the algorithm behind racerTS , which is based on constrained distance geometry. To benchmark the performance of racerTS against CREST and GOAT, we create conformer ensembles for transition states of 20 diverse reactions. To assess the utility of each conformer generator in computational chemistry workflows, we optimize selected low-energy and diverse conformers at the DFT level. We use the generated conformer ensembles and the results of this pipeline to assess conformer generators according to the following metrics: computational cost, exhaustiveness, validity, and accuracy in low-energy regions. Considering the generated ensembles, wefind that racerTS covers the conformer space similarly to CREST and slightly less comprehensively than GOAT, while the validity of the DFT-optimized TSs is better and the accuracy in the low-energy region is sufficient for computational chemistry applications (median error 0.17 kcal/mol). Remarkably, racerTS achieves these results with a significant reduction in required wall-time. Our results thus demonstrate that racerTS is a highly efficient TS conformer ensemble generator, allowing for rapid TS conformer sampling in computational chemistry pipelines. Additionally, racerTS paves the way to create meaningful TS datasets to advance machine learning methods for the discovery of novel and sustainable catalysts.
19 November 2025 in ChemRxiv
The consideration of transition state (TS) conformer ensembles is required to accurately model a reaction, and thus plays a key role in computational catalyst design. While CREST and GOAT are established methods for TS conformer ensemble generation, the associated computational cost remains a major bottleneck in computational chemistry pipelines, including for the generation of large machine learning datasets for catalyst design. To this end, we present racerTS (Rapid and Accurate Conformer Ensembles with Rdkit for Transition States), a method for efficient TS conformer ensemble generation. In this work, we describe the algorithm behind racerTS, which is based on constrained distance geometry. To benchmark the performance of racerTS against CREST and GOAT, we create conformer ensembles for transition states of 20 diverse reactions. To assess the utility of each conformer generator in computational chemistry workflows, we optimize selected low-energy and diverse conformers at the DFT level. We use the generated conformer ensembles and the results of this pipeline to assess conformer generators according to the following metrics: computational cost, exhaustiveness, validity, and accuracy in low-energy regions. Considering the generated ensembles, we find that racerTS covers the conformer space similarly to CREST and slightly less comprehensively than GOAT, while the validity of the DFT-optimized TSs is better and the accuracy in the low-energy region is sufficient for computational chemistry applications (median error $0.17 kcal/mol). Remarkably, racerTS achieves these results with approximate speed-ups of 36x and 4100x compared to CREST and GOAT, respectively. Our results thus demonstrate that racerTS is a highly efficient and accurate TS conformer ensemble generator, allowing for rapid TS conformer sampling in computational chemistry pipelines. Additionally, racerTS paves the way to create meaningful TS datasets to advance machine learning methods for the discovery of novel and sustainable catalysts.
10 December 2024 in ChemRxiv
Determining molecular taste remains a significant challenge in food science. Here, we present FART (Flavor Analysis and Recognition Transformer), a chemical language model capable of predicting molecular taste from chemical structure. Trained on the largest public dataset (15,025 compounds) of molecular tastants to date, FART is the first model capable of parallel predictions across four fundamental taste categories: sweet, bitter, sour, and umami. FART achieves an accuracy above 91% for parallel taste prediction and outperforms previous state-of-the-art binary classifier models that specialize on predicting one taste class. Its transformer architecture allows for interpretability through gradient-based visualization of molecular features. The model identifies key structural elements driving taste properties and demonstrates utility in analyzing known tastants as well as novel compounds. By making both the model and the dataset publicly available, we provide the food science community with tools for rapid taste prediction, potentially accelerating the development of new flavor compounds and enabling systematic exploration of taste chemistry.
19 September 2025 in ChemRxiv
Cubane-type iron-sulfur clusters (Fe4S4) are some of the most versatile metallocofactors and, as such, among multiple functions, primarily responsible for mediating challenging electron transfers (ETs). Their efficient ET chemistry is enabled by a conflated interplay of cofactor-protein interactions, which can be categorized into the covalent first (1°) sphere ones and the non-covalent second (2°) sphere ones. The latter have remained particularly elusive, as they are difficult to observe and assess directly and independently. Accordingly, our understanding of these effects is hampered by their entangled nature. To address this, we herein leverage a systematic series of synthetic Fe4S4 complexes, which allows us to spectroscopically investigate 2° sphere electrostatic interactions and covalent 1° sphere interactions separately from one another. We expand the study of 1° sphere interactions with a histidine-type ligand in [Fe4S4]1+ complexes to the [Fe4S4]2+ and [Fe4S4]3+ oxidation states, supporting the notion that 1° sphere interactions “fine-tune” the electronic/magnetic structure of these systems in a manner that persists at ambient temperatures. In contrast, scrutinizing the 2° sphere electric dipolar interactions in [Fe4S4]1+,2+,3+ complexes revealed that although similar effects are observable at extremely low temperatures, no significant alteration of the clusters’ gross electronic/magnetic structure persists at the temperatures relevant to enzyme function. These results thus not only systematically catalogue the influence of 1° sphere covalent and 2° sphere electrostatic interactions on the observables and properties of Fe4S4 complexes, but also establish a clear energetic distinction between the two. As such, they will facilitate identifying the elusive 2° sphere interactions in biological systems, while also strengthening our biophysical understanding of structure-function relationships in Fe4S4 cofactors.