Physics-based molecular fingerprints from spectral graph theory provide efficient geometry-aware measures of chemical similarity

Jacob W. Toney, Ayleen Y. Farnood, Samir Darouich and Heather J. Kulik (2026)
Highlighted by Jan Jensen

This paper caught my attention for two reasons. First, Morgan fingerprints cannot distinguish stereoisomers of transition-metal complexes, such as cisplatin and transplatin or fac and mer octahedral complexes, despite their potentially very different properties. Second, I often find that Morgan fingerprints tend to assign low similarities to small molecules that I would consider chemically quite similar. A compact fingerprint that incorporates geometry therefore sounds very useful.

The authors construct four matrices from atomic identities and interatomic distances, with entries representing approximate electrostatic, bonding, steric, and dispersion interactions. Each matrix is converted into a graph Laplacian, and 16 selected eigenvalues from each are combined into a fingerprint containing just 64 numbers. Molecular similarity is then calculated from the Euclidean distance between these fingerprints. The method requires neither bond assignment nor molecular alignment, and no training is involved.

The resulting fingerprints distinguish cis/trans and fac/mer isomers, as well as square-planar and tetrahedral coordination geometries. They also distinguish chair and twisted-boat cyclohexane while correctly identifying these as more similar than cyclohexane and piperidine. These examples are encouraging, although distinguishing two structures is only the first requirement: the more interesting question is whether the resulting distances reflect useful chemical similarity. Also, because the fingerprints depend only on interatomic distances, they cannot distinguish mirror-image enantiomers.

Figure 4 in the paper provides a more demanding test. Each molecule’s property is estimated by averaging the values for its five nearest neighbours in fingerprint space. For the small organic molecules in QM9, spectral fingerprints perform especially well for atomisation energies and polarisabilities, but, surprisingly, is tied with Morgan fingerprint for dipole moments and HOMO-LUMO gaps. 

What surprised me even more is that Morgan fingerprints outperform the spectral fingerprints for both dipole moments and HOMO–LUMO gaps of the transition-metal complexes. You might expect the explicit geometry information to be particularly valuable here. One complication is that Morgan fingerprints are evaluated only for the roughly 73% of complexes that RDKit can parse, whereas spectral fingerprints work for the entire dataset. Nevertheless, the result is a useful reminder that greater structural resolution does not automatically produce more useful property neighbourhoods.

The computational cost is also impressive: the authors report generating fingerprints for 2.7 million protein-fragment structures in about three minutes, once coordinates are available. The code is available as SpectralScore. This approach might thus also be a cheaper alternatives to the shape-similarity algorithms used in drug discovery. 

This work is licensed under a Creative Commons Attribution 4.0 International License.

Modeling Reaction Feasibility to Identify New Opportunities for Sigmatropic Rearrangements

Jordan P. Liles, Nicholas Casetti, Thijs Stuyver, Connor W. Coley (2026)
Highlighted by Jan Jensen

One thing I tend to go on and on about the is the dearth of prospective experimental verification of computational predictions, so this study is a welcome and interesting addition.

The authors combine a robust TS optimizer with a reactive MLIP that can efficiently compute barriers for sigmatropic rearrangement and use the methodology to screen for feasible reactions. 

For example, they targeted natural products from the COCONUT database that lack reported synthetic preparations. Beginning with the full database, they applied a set of retrosynthetic rearrangement templates to enumerate precursors capable of forming natural products through a single hypothetical rearrangement. They used ASKCOS to identify precursors accessible in three or fewer synthetic steps. Structural complexity was assessed using the synthetic accessibility score, and kinetic feasibility was defined by a barrier-height threshold of 40 kcal/mol. This workflow yielded 23 promising candidates and their corresponding rearrangements. Among these candidates, they identified (±)-norbrachycoumarin. Although it was first isolated in 1986 it had never been chemically synthesized. They subsequently confirmed the predicted rearrangement experimentally!

More generally, prospective, physics-based evaluation of reaction mechanisms could accelerate synthetic chemistry. Analyzing kinetically relevant elementary steps can inform the feasibility and selectivity of key transformations while revealing useful retrosynthetic disconnections that might otherwise be overlooked. However, accurate reaction modeling remains challenging because of factors such as solvation, transition-state conformational diversity, catalysis, open-shell species, and metal-containing systems. Mechanistic uncertainty presents an additional difficulty, as reliable predictions require competing reactions and off-cycle pathways to be considered.

The authors conclude:

The general strategy of applying physics-driven mechanism evaluation prospectively holds great promise for accelerating synthetic chemistry. Probing kinetically relevant elementary steps can help evaluate the feasibility and selectivity of key reactions while planning synthetic routes as well as help reveal strategic disconnections that may go overlooked in initial route designs. However, several challenges in reaction modeling remain, including solvation, conformational ensembles of transition states, catalysis, open-shell systems, and heavier elements such as metals. Equally complicating can be mechanistic uncertainty, wherein accounting for competing side reactions and/or off-cycle pathways is crucial for oracle accuracy. Nonetheless, our results demonstrate that current methods are, at present, capable of delivering actionable predictions at a meaningful scale for well-behaved reaction classes.

This work is licensed under a Creative Commons Attribution 4.0 International License.

Rapid Generation of Transition-State Conformer Ensembles via Constrained Distance Geometry

Stefan P. Schmid, Henrik Seng, Thibault Kläy, and Kjell Jorner (2026)
Highlighted by Jan Jensen

Here's another example of a tool that has made research life in the group significantly easier. racerTS takes a single TS structure and generates an ensemble of TS conformers. While this could already be done with CREST or GOAT, those methods were too expensive for the workflows we are developing in the group, so we basically ignored the problem. For minima, we used RDKit for conformational sampling, and we needed something similar for TSs—and that is exactly what racerTS provides.

racerTS is an RDKit-based pipeline for rapidly generating transition-state conformer ensembles using constrained distance geometry. It requires an XYZ structure of a TS guess, the total molecular charge, and the atom indices defining the reaction center; reactant or product SMILES may optionally be supplied. The coordinates and charge are converted into an RDKit Mol object. The reaction-center atoms and their automatically identified nearest neighbors collectively define the “frozen atoms.” ETKDG then generates conformers by repeatedly sampling a distance-bounds matrix while preserving the reaction-center geometry through positional and distance constraints. The resulting structures are refined with MMFF94, with UFF used as a fallback for unsupported atom types. Duplicate conformers are removed when their heavy-atom RMSD relative to a lower-energy conformer is below 0.125 Å. Structures more than 20 kcal/mol above the force-field minimum are discarded, and the surviving ensemble is written to a multi-structure XYZ file. For improved energetic ranking, the full workflow can additionally optimize the ensemble with GFN-FF while fixing the frozen atoms, rank the conformers using GFN2-xTB single-point energies, and repeat the pruning with a 6 kcal/mol window.

Across 20 diverse reactions, racerTS was approximately four times faster than CREST and 500 times faster than GOAT on a single CPU core. This comparison used GFN2-xTB//GFN-FF for racerTS and CREST, whereas GOAT was run at the more expensive GFN2-xTB level. racerTS, CREST, and GOAT all gave median activation-energy errors below 0.18 kcal/mol, with racerTS yielding an error of 0.17 kcal/mol. The xTB post-optimization step was important for racerTS’s accuracy; omitting it increased the speed but degraded the energy ranking. Overall, racerTS offers accuracy and conformational coverage comparable to those of CREST—though it is somewhat less exhaustive than GOAT—at a substantially lower computational cost, with occasional larger energy outliers for individual reactions.

This work is licensed under a Creative Commons Attribution 4.0 International License.

From SMILES Codes for Reactants and Products to Transition States With VeloxChem

Bastiaan van Hoorn, Patrick Norman and Mårten S. G. Ahlquist (2026)
Highlighted by Jan Jensen

Imagine if you could just draw a reaction in, say, ChemDraw, press a button, and get the TS (like shown above). Well now you pretty much can.**

Many algorithms construct TS guess structures by interpolating between reactant and product structures and therefore require that the atoms are in the same order. This can be very tedious to set up by hand and makes inclusion into automated workflows difficult.

In addition, I've found that the success of interpolation methods for biomolecular reactions depends heavily on how the reactant and product structures are aligned. Some methods try to solve this by rigid alignment, but that tends not to be sufficient for large, flexible molecules.

This paper offers automatic atom mapping and alignment, which is a massive step forward, in addition to a novel FF-based interpolation method that even includes some conformational sampling. And it works for "real-world" systems as you can see above.

Although VeloxChem can also optimise TSs, you can also interface it with other packages though a Python API. Here you can extract the interpolation path from VeloxChem (see this example notebook) and use that info (e.g. the TS guess or the aligned, atom-mapped reactant and product coordinates) as input to other packages if you want. 

This is a really great tool that we are already using quite a bit in my group.

**You have to provide SMILES strings, which you can easily get from ChemDraw. For transition metal complexes (TMCs) you have to give (non-mapped) coordinates instead, because VeloxChem uses RDKit. ChemDraw can't produce RDKit-compatible SMILES for TMCs (but Schrödinger Sketcher can). Also, the 3D structures that RDKit produces for TMCs are not very good. However, my group and others are working on fixing this, so that should be available soon. 

This work is licensed under a Creative Commons Attribution 4.0 International License.

Developing Pharmaceutically Relevant Pd-Catalyzed C−N Coupling Reactivity Models Leveraging High-Throughput Experimentation

Seung Kyun Ha, Dipannita Kalyani, Michael S. West, Jessica Xu, Yu-hong Lam, Thomas Struble, Spencer Dreher, Shane W. Krska, Stephen L. Buchwald, and Klavs F. Jensen (2025)
Highlighted by Jan Jensen

Yield prediction is one of the most difficult and important challenges for machine learning applied to chemistry. This paper is a useful contribution because it provides a relatively large and systematic high-throughput dataset of ca. 4000 Pd-catalyzed C−N coupling reactions, spanning a wide variety of secondary amines and aryl bromides relevant to medicinal chemistry.

One important caveat is that the study does not address full reaction-condition optimization. All scope reactions are run using a single set of reaction conditions: one catalyst, one base, and one solvent system. The task is therefore better described as substrate-scope prediction under fixed conditions, rather than general prediction of reaction yield across arbitrary reaction conditions.

Significantly, the authors provide a useful reality check on the quality of yield data. For 32 repeated reactions, the measured product Liquid Chromatography Area Percent (LCAP) values correlate poorly, with (R^2 = 0.35). This experimental variability motivates their decision to treat the problem as binary classification rather than regression. A threshold of 20% product LCAP is chosen to define a “successful” reaction, and the repeated reactions are then consistent under this classification scheme in 27 cases. This supports a broader cautionary point: if yields from carefully controlled HTE experiments are already noisy at the level of absolute values, then predicting precise yield values from heterogeneous literature or web-scraped data is likely to be extremely difficult, and perhaps unrealistic in many settings.

The authors construct four different test sets to ascertain whether ML models can be used to extrapolate to unseen amines (amine OSS), aryl bromides (ArX OSS), or both (Both OSS) in addition to standard interpolation (DRS(n) where n is the percentage of the dataset used for training).

The authors compare several model classes and molecular representations, including random forests, decision trees, AdaBoost, fully connected neural networks, and MPNNs using Chemprop. Input features include one-hot encodings, Morgan fingerprints, quantum-mechanical fingerprints, molecular graphs, and combinations of these. Overall, the best models are usually either random forests with fingerprint-based descriptors, sometimes augmented with QM descriptors for the reacting components, or MPNNs. However, the optimal model and representation depend on the data split, which is itself an important result: there is no single universally best model for all generalization tasks.

The best model for each split is then used to design a corresponding prospective validation library of 96 reactions. The models are first retrained on the full experimental dataset using the best architecture, input features, and hyperparameters identified from the retrospective modeling. For the DRS validation library, the DRS25 settings are used. Each validation library is constructed so that approximately half of the reactions are predicted to give >20% LCAP and half are predicted to give <20% LCAP. The confidence threshold is >0.9 for the Amine OOS, ArX OOS, and DRS libraries, and >0.8 for the Both OOS library. For OOS amines or aryl halides, the selected substrates must also have a maximum Tanimoto similarity <0.7 to the corresponding substrates used in the model-building dataset. Thus, the validation libraries are not random samples of chemical space; they are enriched for reactions where the model is sufficiently confident.

The prospective validation results are impressive. For the Amine OOS library, the RF model gives 11 false positives, and no false negatives. For the ArX OOS library, the MPNN gives 3 false positives, and 2 false negatives. For the Both OOS library, the RF model performs less well but still gives useful enrichment, with most errors arising from false positives rather than false negatives. For the DRS25 library, the RF model performs extremely well, with essentially perfect precision and only one false negative. Overall, the models are especially good at avoiding false negatives, which is important in a medicinal chemistry setting because false negatives could cause chemists to discard reactions that would actually work.

Having said that, this study represents something close to a best-case scenario for reaction-outcome prediction. The dataset is large by the standards of synthetic chemistry, with around 4000 systematically generated reactions. The reactions are all run under the same conditions, reducing experimental heterogeneity. The positive rate is also relatively high: about 35% of the reactions exceed the 20% LCAP threshold. This makes the classification task easier than many realistic discovery settings where successful reactions are much rarer. Finally, because the dataset is large and the hit rate is high, the models can make a substantial number of high-confidence predictions, which enables the construction of balanced validation libraries with 50% predicted successes and 50% predicted failures. In smaller, noisier, or more imbalanced datasets, this level of prospective performance would likely be much harder to achieve.

This work is licensed under a Creative Commons Attribution 4.0 International License.

Harnessing AtomisticSkills for Agentic Atomistic Research

Bowen Deng, Bohan Li, Matthew Cox, Hoje Chun, Juno Nam, Artur Lyssenko, Sathya Edamadaka, Jurgis Ruza, Xiaochen Du, Nofit Segal, Jesus Diaz Sanchez, Mingrou Xie, Ty Perez, Yu Yao, Miguel Steiner, Sauradeep Majumdar, Charles B. Musgrave III, Anirban Chandra, Abhirup Patra, Detlef Hohl, Connor W. Coley, Ju Li, Rafael Gómez-Bombarelli (2026)
Highlighted by Jan Jensen

AtomisticSkills is a hierarchical research framework in which skills encode reusable, mid-level scientific workflows, while tools provide low-level, type-checked computational operations that agents can reliably call to execute those workflows. While LLMs can in principle assemble such workflows from package documentation and first principles, in practice their performance degrades as context length grows. Put another way, trying to keep the manuals and execution details for RDKit, ORCA, and related tools in context at the same time is likely to increase hallucinations in the proposed workflow. Instead, complicated workflows are distilled by experts into SKILL.md files that outline how the tools are to be used, and these can be loaded into general-purpose coding agents such as Claude Code and Codex.

I especially like this last point. AtomisticSkills lets researchers use a tool they may already be familiar with, but apply it to new scientific problems. It looks like an interesting way to share robust workflows with non-experts. Take, for example, the installation of AtomisticSkills  itself: it basically amounts to downloading the repository and telling Codex to “Install AtomisticSkills according to its docs/setup.md guide,” after which the agent interactively guides the user through comparatively complicated steps such as creating environments, configuring API keys, and registering MCP servers.

For example, while we have made the xTB version of our EsNuEl workflow available through a web server, making the DFT versions available there was not practical. Installing it locally from the repo is of course possible, but perhaps a little intimidating for the target group of synthetic chemists. An approach like this might be more palatable: package the workflow as a skill, provide tested scripts and examples, and let a general-purpose coding agent guide the user through local setup and execution.

This work is licensed under a Creative Commons Attribution 4.0 International License.

Density Functional Theory Surrogate Enables Fast and Broad Computational Evaluation of Homogeneous Transition Metal Catalytic Energy Landscapes

Kevin P. Quirion, Wang-Yeuk Kong, Britton Stanley, Jyothish Joy, and Daniel H. Ess (2026)
Highlighted by Jan Jensen

It has been about 10 months since Meta FAIR released the Universal Models for Atoms, or UMA, machine-learning interatomic potentials. Since then, the first independent benchmarking studies have begun to appear, and this paper by Quirion and co-workers asks a very practical question: can UMA be used as a fast surrogate for DFT in homogeneous organometallic catalysis?

The authors examine seven catalytic/organometallic case studies taken from the literature, including Ir pincer alkane dehydrogenation, Rh hydroformylation, Ru olefin metathesis, Pd Buchwald–Hartwig amination, Cu-catalyzed difluorocarbene insertion, Ni asymmetric radical capture/reductive elimination, and a dinuclear Ni–Ni naphthyridine-diimine cycloaddition.

For literature geometries, they recompute reaction energies using ωB97M-V/def2-TZVPD single points, which is close to the level of theory that UMA is trained to reproduce. They then compare these values to UMA-S and UMA-M single-point energies, and in many cases also to UMA-optimized structures and energies. 

The headline result is encouraging: in most cases, UMA tracks ωB97M-V very well, often within a few kcal/mol and with good agreement in relative barriers and reaction-profile shapes. This is particularly impressive because the systems include different metals, oxidation-state changes, large ligands, charged species, and transition states. For routine conformer screening, preliminary mechanism mapping, or fast evaluation of many candidate catalysts, this suggests UMA could be genuinely useful.

There are, however, two important problem cases.

The first is the Cu-catalyzed difluorocarbene insertion, where the key issue is an open-shell singlet intermediate. UMA could not locate the TS1e transition state during optimization or NEB, gave unphysical conformational changes when optimizing the singlet 3e, and predicted the triplet state of 3e to be much lower than the singlet. At first glance this looks like a UMA failure, but ωB97M-V itself has similar problems with the singlet–triplet energetics. So this is not simply a machine-learning-potential problem. UMA is trained to reproduce ωB97M-V-like energies and forces; it should not be expected to magically repair failures of the underlying DFT reference method. The more specific concern is that UMA also has practical difficulties optimizing the open-shell singlet surface and locating the associated transition state. It was not tested whether ωB97M-V had the same problem.

The second problem case is the dinuclear Ni–Ni naphthyridine-diimine diene cycloaddition. Here UMA struggles with the relative spin states and barriers. In particular, it does not reproduce the same doublet/quartet ordering as ωB97M-V, and it overstabilizes some parts of the profile. This is perhaps less surprising because OMol25 did not include multinuclear transition-metal complexes, and the authors note that the naphthyridine-diimine ligand is not represented in the training set. Interestingly, the optimized geometries are not disastrous: UMA-S gives heavy-atom RMSDs of roughly 0.22 Å for the doublet and 0.36 Å for the quartet relative to the reported M06-L structures. So the failure is more severe for relative energetics and spin-state ordering than for generating plausible structures.

Overall, the study is a strong endorsement of UMA as a practical tool for organometallic mechanism work, provided it is used with the same caution one would apply to DFT. UMA appears especially promising for rapid conformer screening, approximate reaction-profile generation, and preoptimization before higher-level single-point calculations.

One unresolved issue is training-set overlap. The authors write that the OMol25 training database is so large that it “cannot be easily queried,” and that UMA does not provide an intrinsic nearest-neighbor or structure-comparison analysis for new inputs. That is a real limitation: if a benchmark system, or something very close to it, is already in the training data, the benchmark is much less informative about out-of-distribution generalization.

At the same time, the paper also states that the authors queried the dataset for the naphthyridine-diimine ligand and provide code in the Supporting Information. So the situation is somewhat unclear. The database may be inconvenient to search, but it does not seem impossible to search. For future UMA benchmark studies, it would be very useful to include at least a basic training-set check: for example, filtering OMol25 by metal, composition, charge, spin state, ligand identity, and local coordination environment. This would help distinguish cases where UMA is genuinely extrapolating from cases where it is interpolating within a familiar chemical neighborhood.

This work is licensed under a Creative Commons Attribution 4.0 International License.