Papers for

pharmaceutical chemists

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

Latent diffusion model guides small-molecule design for drug properties

Classifier-Guided Latent Diffusion for Small-Molecule Generation

Abstract: Diffusion models have emerged as a promising class of generative models for molecular design because of their ability to learn complex chemical distributions and support conditional or guided generation. In this work, we investigate a latent diffusion pipeline for property-guided small-molecule generation. The proposed framework represents canonicalized SMILES strings using a frozen ChemBERTa encoder, trains a Transformer-based decoder to reconstruct SELFIES strings from molecular latent embeddings, and learns an unconditional denoising diffusion model in the ChemBERTa latent space. To enable property-directed generation, a differentiable latent property predictor is used to guide the diffusion sampling process. Generated SELFIES strings are converted to SMILES, validated using RDKit, and evaluated using molecular validity, uniqueness, novelty, and predicted property scores. We study guidance toward two drug-relevant binary endpoints: blood--brain barrier permeability and inhibition of human $β$-secretase 1. The results suggest that latent diffusion over pretrained molecular embeddings can generate chemically valid and novel molecules while enabling controllable steering toward desired molecular properties.

Fri 13 NovMachine Learning
The gist
Designing new small molecules with specific drug-like properties is challenging. The authors built a system that generates molecule structures by learning from existing chemical data and guiding the design toward desired traits like crossing the blood-brain barrier or inhibiting a target protein. They use a machine learning approach called latent diffusion combined with molecular sequence representations, allowing control over the properties of the generated molecules. This method helps create valid, novel chemical structures tailored to certain biological effects.
Open → 2611.17650v1

Chemical reactions simulated by tracking electron movements precisely

Mechanism-Level Chemical Reaction Simulation with Electron Bookkeeping Transformer

Abstract: Modeling chemical reaction mechanisms is central to understanding and predicting chemical reactivity. We present Electron Bookkeeping Transformer (Bookkeeper), a data-driven model that simulates chemical reactions at mechanistic resolution by generating electron movement trajectories. Following the symbolic arrow-pushing formalism in chemistry, Bookkeeper explicitly models each electron as a token and predicts its movement sequentially, ensuring mass and charge conservation during both elementary step and reaction pathway predictions by construction. On reaction-size-balanced test sets, Bookkeeper improves top-1 elementary-step accuracy, top-1 pathway rank accuracy, and top-1 to top-5 product prediction accuracy over FlowER. On the entire test set, it achieves higher elementary-step top-1 accuracy but lower pathway accuracy than FlowER. Bookkeeper also samples elementary steps substantially faster, adapts to several unseen reaction types with minimal fine-tuning, and generates candidate chemical transformations for further screening. These results suggest that explicit electron bookkeeping is a useful framework for virtually screening chemical reactions across diverse reagents, validating computer-assisted synthesis planning predictions, and providing mechanistic interpretations of reactivity.

Thu 12 NovComputational Engineering, Finance, and Science
The gist
Predicting how chemicals react often relies on understanding tiny electron shifts. The authors created a model that treats each electron like a moving token, following chemistry’s arrow-pushing rules. This model, called Bookkeeper, can predict reaction steps with good accuracy, keep track of charge and mass, work faster, and adapt to new reaction types. It helps chemists understand reactions in detail and suggests new ways to test chemical transformations.
Open → 2611.17548v1

Machine learning boosts chemical structure guesses using reaction context

Context-Aware Automated NMR Structure Elucidation

Abstract: Machine learning approaches are rapidly revolutionizing the ability to analyze and extract structural insights from chemical data, enabling high-throughput chemical discovery. While existing data-driven approaches have shown remarkable promise for predicting chemical structures in the de novo setting, incorporating additional chemically informed context offers the opportunity to further improve these models' performance and extend their real-world utility. Here we provide a straightforward protocol to combine artificial-intelligence-based spectroscopic elucidation pipelines with contextual chemical data, e.g., known reactants or a desired product, by using automated retrosynthetic analysis to generate large training sets of molecules and contextual data in the form of synthetic precursors. We demonstrate our approach by showing that our recently introduced NMR structure elucidation framework can be trained to effectively leverage molecular precursor context. We show that when the chemical context provides useful information on the target molecule we obtain dramatically improved accuracy over using NMR spectra alone, obtaining Top-15 accuracies of up to 95.4% for molecules up to 40 heavy atoms spanning the full elemental range of organic chemistry. Furthermore, we demonstrate that our approach is highly robust, such that even when the context provided is uncorrelated with the target, one still obtains accuracies commensurate with those of the original NMR structure elucidation framework. This approach provides an efficient route to improve a wide range of spectroscopic structure elucidation platforms using contextual data.

Wed 4 NovMachine Learning
The gist
Figuring out the structure of molecules from NMR data is tricky. The authors show that using clues about how molecules are made (their precursors) helps AI do a much better job at guessing these structures. Even if these clues are unrelated, the system still performs well. This makes identifying chemical structures faster and more accurate.
Open → 2611.07133v1

Reinforcement learning improves small molecule identification with fewer tests

From MS/MS evidence to adaptive spectroscopic characterization: Reinforcement learning guided by candidate molecular structures

Abstract: Small-molecule identification from tandem mass spectrometry (MS/MS) may require complementary measurements. Herein, we propose a structure-informed deep Q-network (SI-DQN) that uses existing MS/MS evidence to decide whether further characterization is needed, which measurement to request, and when to stop. The policy combines spectral representations and candidate rankings with score-weighted descriptors of the candidates' chemical features and structural differences. The policy was evaluated on 79,462 held-out samples with simulated MS/MS, infrared (IR), and nuclear magnetic resonance (NMR) spectra from the Multimodal Spectroscopic Dataset. Evaluation used fixed hard candidate sets containing the correct structure and equal normalized acquisition costs. Averaged across three training runs, SI-DQN increased Top-1 accuracy from 80.10% to 93.78% for positive-ion MS/MS starts and from 70.44% to 93.31% for negative-ion starts. The corresponding mean additional measurement counts were only 0.227 and 0.340 per sample, respectively. Complete acquisition achieved 97.72% accuracy with four additions, whereas SI-DQN reduced the additional measurement count by 94.3% and 91.5%. Relative to a rule-based policy, SI-DQN reduced the composite loss combining terminal ranking error and acquisition cost by 19.3% and 25.5%. Adding structural information to DQN further improved accuracy while reducing requests across all three training runs. At comparable ranking uncertainty, SI-DQN increased request rates relative to DQN more strongly in groups with high structural disagreement, where larger reductions in composite loss were also observed. Together, these results support using the ambiguity among candidate structures after MS/MS analysis to guide further characterization, providing a quantitative basis for balancing identification performance against subsequent measurement effort.

Thu 29 OctMachine Learning
The gist
Identifying small molecules using mass spectrometry often needs extra tests to be sure about the result. The authors developed a method that uses machine learning to decide if more measurements are needed, which to do, and when to stop. Their approach smartly combines chemical information and measurement data to improve accuracy while reducing extra tests. This method was tested on many simulated examples and showed it can identify molecules more accurately with far fewer additional tests than traditional methods.
Open → 2610.40264v1

Controllable molecular optimization improves drug lead properties efficiently

ReMedI: Inference-Time Refinement and Self-Distillation for Controllable Molecular Optimization

Abstract: In drug development, lead optimization must improve competing properties, such as increasing permeability and reducing toxicity, while limiting scaffold changes that could compromise the lead's potency. RL-based post-training, such as C-MORAL, addresses this problem by fine-tuning molecular language models to directly optimize molecular leads for objectives of interest, which relies on expensive task-specific datasets and faces challenges in achieving multi-objective optimization (MOO). We argue a pre-trained model may already contain useful molecular edits that zero-shot generation cannot efficiently reach. Motivated by the limited search capabilities of zero-shot generation, we develop an efficient Metropolis-Hastings search strategy to better explore the proposal distribution. We propose ReMedI, which uses the molecular language model as a base proposer for inference-time alignment. Given a lead and a set of multi-objective tasks, the model proposes edits conditioned on the current molecule, while Metropolis--Hastings refinement guides sequential edits under an objective for property improvement, preservation, and lead similarity. On both C-MuMOInstruct and SMDD benchmarks, refinement improves frozen base models across three backbones, reaching significantly higher multi-objective optimization success rates compared to baselines. These search traces can then distilled into the base model which improves zero-shot generation and subsequent search using approximately 7.8 times less training molecular leads than that of C-MORAL. ReMedI achieves approximately 2.8 times the strict-success coverage of RL baseline C-MORAL, including on out-of-distribution tasks, with matched lead-relative similarity.

Fri 23 OctMachine LearningArtificial Intelligence
The gist
Improving drug molecules is tricky because changes can help some properties but harm others. The authors created a method called ReMedI that refines molecular designs step-by-step during inference time, making better improvements without needing lots of new training data. They use a smart search to tweak molecules while keeping them similar to the original and meeting multiple goals. This method works better than previous systems and learns to generate improved molecules more efficiently afterward.
Open → 2610.32103v1

Chemistry tool uses AI to plan complex molecule synthesis routes

ReTRACE: Feedback-Guided LLM Agent in a Chemistry Harness for Retrosynthesis Planning

Abstract: Computer-aided retrosynthesis has advanced through single-step prediction, tree search and route scoring, but complex targets still require judgement over the whole route, where stereochemistry, functional-group compatibility, cost and greenness interact. Here we present \textbf{\mbox{ReTRACE}} (\textbf{RE}trosynthesis with \textbf{T}ool-orchestrated \textbf{R}easoning \textbf{A}nd \textbf{C}ontext-aware \textbf{E}vidence), an agent that places a large language model (LLM) in a chemistry harness that integrates molecular tools, complementary single-step prediction engines, a purchasable starting-material library, a route memory and a multi-criterion route evaluator. Routes are judged by a new metric that combines programmatic checks with adversarial LLM review, and its verdicts are compared with a synthetic chemist's. The agent learns continually from evaluator feedback during search, diagnosing weak routes and backtracking to revise them. With Claude Opus 4.8, at least one of the top 10 routes proposed by ReTRACE passed the audit for 87.4\%, 95.0\% and 86.7\% of targets on the three benchmarks. Two prospective wet-lab syntheses show that ReTRACE route proposals can be converted into executable procedures after expert revision.

Sat 17 OctArtificial Intelligence
The gist
Planning how to make complex molecules is hard because many factors like cost and safety must be considered across multiple steps. The authors created ReTRACE, a system that combines a large language model with chemistry tools to suggest full synthesis routes. It checks these routes using automated tests and expert-like reviews, learning to improve suggestions over time. The system showed strong success on standard benchmarks and was tested in real lab experiments after expert refinement.
Open → 2610.24744v1

GEMS improves small molecule identification using learned editing of graphs

GEMS: Molecular Structure Elucidation as Learned Search over Isomer Graphs

Abstract: Identifying small-molecule structures from molecular formulas and tandem mass spectra remains a bottleneck in untargeted metabolomics and natural-product discovery. Even when database retrieval or de novo prediction misses the exact structure, the resulting candidates can provide starting points for exploring nearby structures through a few bond edits. We introduce GEMS (Guided Editing of Molecular Structures), which refines candidates through learned 2-switch bond edits that preserve the molecular formula and each atom's total valence. GEMS uses MIST, a pretrained spectrum-to-fingerprint model, to predict fingerprint probabilities once per spectrum. A neural policy compares each candidate's fingerprint with these predictions, links present features to the atoms and bonds that generate them, and scores all valid edits in one forward pass. The policy is trained on shortest edit paths between structures, so it can be pretrained on PubChem isomers without spectra and then fine-tuned with spectrum-predicted fingerprints. At inference, GEMS searches from multiple initial candidates matching the supplied molecular formula and ranks candidates by fingerprint agreement and multi-start consensus. With the target structure excluded from PubChem initialization, three edit rounds achieve Hit@1/Hit@10 of 29.64%/50.81% on NPLIB1 and 30.22%/41.33% on MassSpecGym, exceeding published baselines under comparable settings. On NPLIB1, GEMS achieves higher identification accuracy in about 3.3 minutes end to end, compared with 26.2 hours for our FRIGID run. GEMS also improves identification from generated candidates, including those refined by FRIGID. Experiments with exact target fingerprints identify fingerprint prediction error as a bottleneck in target recovery and final ranking. Deeper search recovers additional targets without retraining, with larger gains from weaker initial candidates.

Fri 16 OctArtificial Intelligence
The gist
Identifying the exact structure of small molecules just from their formulas and mass measurements is very hard. The authors created GEMS, a method that starts with possible candidates and improves them step-by-step by changing bonds while keeping the overall formula correct. GEMS learns how to make these changes by training on known molecule variations and predicting important molecular features from the mass data. This method finds better matches faster than previous approaches and can also improve candidates generated by other tools.
Open → 2610.23617v1

Fragflex generates valid and unique molecules by adding chemical fragments

FragFlex: Fragment-Based Flexible Molecular Generation for Exploring Vast Chemical Space

Abstract: Drug discovery requires efficient exploration of an astronomically large chemical space. Molecular graph diffusion models offer a promising approach to this task, but many require graph size to be fixed before generation. Although atom-level insertion and deletion can enable flexible-size generation, jointly modeling molecular size, atom types, and bonds while maintaining chemical validity remains challenging. We introduce FragFlex, a flexible-length discrete diffusion model that performs node insertion and deletion at the fragment level. During denoising, FragFlex jointly learns the size, fragment identities, and inter-fragment connectivity of the fragment graph. A neural attachment-site predictor and valence-aware assembler then convert the generated fragment graph into an atom-level molecule. In large-scale experiments generating 10 million molecules, FragFlex achieves near-perfect chemical validity while maintaining high uniqueness, demonstrating a strong balance between the two.

Sun 11 OctMachine Learning
The gist
Discovering new drugs is hard because there are so many possible molecules to try. The authors created FragFlex, a method that builds molecules by adding and removing parts called fragments instead of single atoms. This method learns how to join fragments correctly and makes sure the molecules make chemical sense. Their tests show FragFlex can create many unique and valid molecules, helping explore new chemical possibilities efficiently.
Open → 2610.15284v1

Unified model and benchmark connect enzymes to reactions and molecules

Elucidating the Space of Enzymatic Reaction: A Unified Benchmark and Pretrained Model

Abstract: Existing reaction models primarily learn molecular transformations, whereas enzy- matic reactions depend jointly on molecular structure and catalytic function. We formulate this problem as learning an enzymatic reaction space linking reactants, products, and Enzyme Commission (EC) annotations. To characterize this space, we introduce VenusRX-Bench, a unified benchmark for forward reaction prediction, single-step retrosynthesis, and EC-number prediction. VenusRX-Bench integrates reactions from multiple biochemical databases with standardized curation, leakage- controlled splits, and consistent evaluation. Benchmarking representative chemical and enzymatic models reveals a clear chemical-to-enzymatic domain gap, driven by limited domain data, catalytic-context dependency, and the difficulty of modeling large biomolecular structures. To bridge this gap, we develop VenusRX, a unified T5-style sequence-to-sequence model for enzymatic reactions. VenusRX jointly learns forward prediction, ret- rosynthesis, and reaction reconstruction, with two-stage training on millions of template-expanded reactions followed by real biochemical reactions. In addition, optional EC conditioning incorporates catalytic context, while Molecule Library- Constrained Decoding improves the generation of complex biomolecules. Across benchmark tasks and challenging generalization splits, VenusRX achieves the best or competitive performance on most evaluated settings over representative chem- ical and enzymatic baselines. Moreover, EC information consistently improves reaction prediction, while learned reaction representations support accurate EC prediction, revealing a bidirectional relationship between reaction structure and catalytic function. Together, VenusRX-Bench and VenusRX provide a unified framework for elucidating and modeling enzymatic reaction space

Thu 8 OctArtificial Intelligence
The gist
Predicting chemical reactions involving enzymes is hard because enzymes change molecules depending on their specific function, not just their structures. The authors created a benchmark called VenusRX-Bench that combines data from several biochemical sources to test how well models predict enzymatic reactions, the starting materials needed, and enzyme types. They also made a new computer model called VenusRX that learns from millions of reactions and includes information about enzyme function to improve accuracy. Their work helps clarify how enzyme types and reactions relate, making it easier to study and predict complex biochemical processes.
Open → 2610.11694v1

Molecular design improved by searching continuous synthesis routes space

Navigating Route Latent Space for Synthesizable Molecular Design

Abstract: Goal-directed molecular design has advanced rapidly, yet a substantial proportion of designed molecules remain difficult to synthesize in practice, limiting their real-world utility. Prior synthesizability-aware methods either project generated molecules back to synthesizable analogs that deviate from the intended target, or optimize directly in discrete synthesis spaces that lack a continuous landscape for efficient search. We argue that this limitation mainly comes from the search space rather than the optimizer. To address this, we propose RouteFlow, a framework that reformulates synthesizable molecular design as a search over a continuous route latent space, where each latent maps back to a complete synthesis route and synthesizability is inherently preserved. To navigate this space, we adopt reward-guided flow matching as an efficient sampler that steers toward high-property regions. Since reward optimization may push latents off the manifold of real synthesis routes, where decoding becomes unreliable, we further introduce a cycle-consistency mechanism to stabilize fine-tuning. Across 16 optimization tasks from Therapeutic Data Commons, RouteFlow achieves the best sample efficiency among synthesizability-aware baselines, with the best synthetic accessibility and the highest retrosynthesis success rate. Our results also confirm that the proposed cycle-consistency reliably keeps optimization on-manifold while improving target properties, supporting effective synthesizable molecular discovery.

Tue 6 OctArtificial IntelligenceComputational Engineering, Finance, and Science
The gist
Designing new molecules often hits a roadblock because many are hard to make in real life. The authors found that searching in the usual way misses the smooth paths that connect easy-to-make molecules. They created RouteFlow, a method that looks through a continuous space of possible synthesis plans, making sure the molecules can actually be built. This helps find molecules that are both useful and practical to produce.
Open → 2610.07560v1

Interpretable fast model predicts solubility in water and organic solvents

DISSOLVR: An Interpretable and Fast Framework for Aqueous and Organic Solubility Prediction

Abstract: High-fidelity solubility prediction is fundamental to pharmaceutical development and environmental partitioning, where accurate modeling must couple molecular structure with thermodynamic behavior across diverse chemical environments. However, recent advancements have been dominated by deep learning architectures that often sacrifice physical interpretability for predictive power. We challenge this trend by showing that state-of-the-art performance does not require such non-transparent architectures. To address this, we introduce DISSOLVR, a transparent framework for molecular solubility prediction. In addition, we perform a comprehensive literature review and a benchmarking study against various methods. We show that DISSOLVR approaches the aleatoric limit of experimental uncertainty and achieves OOD generalization through structural invariance, derived by mapping molecules to physically-grounded descriptors. Then, we present an LLM-assisted post-hoc explanation pipeline that bridges the gap between symbolic model artifacts and chemically grounded narratives. Finally, a comparative benchmark of a survey involving 22 expert chemists reveals that expert evaluators provide deep insights.

Thu 1 OctMachine Learning
The gist
Accurately predicting how well chemicals dissolve in liquids is very important for making medicines and studying the environment. Many recent methods use complex AI models that are hard to understand. The authors introduce DISSOLVR, a transparent and quick method that uses understandable chemical properties to predict solubility nearly as well as the best existing models. They also create a way to explain these predictions in simple chemical terms with help from language models. Experts reviewed their approach and provided valuable feedback on its insights.
Open → 2610.02574v1

Generative model predicts molecule breakups for drug design

RetroGEF: Dynamic Graph Edit Flow for Single-Step Retrosynthesis

Abstract: Retrosynthesis enables the discovery of viable synthetic routes to target molecules. It plays a central role in modern drug discovery and materials design. Retrosynthesis involves molecular graph transformations that can change both connectivity and graph size. These transformations may introduce reactant components absent from the target while revising the product-derived structure. To model these transformations, we propose RetroGEF, a flow-based generative model for single-step retrosynthesis. Starting from the target molecule, it constructs possible reactants by adding atoms and changing bonds in the molecular graph. RetroGEF models molecular transformations and changes in graph size within the same generative process, rather than relying on a fixed-size graph canvas. It learns this process directly from product--reactant pairs without requiring a prescribed edit order. Experiments on representative retrosynthesis benchmarks demonstrate that RetroGEF achieves state-of-the-art performance.

Tue 29 SeptMachine Learning
The gist
Designing new drugs often involves figuring out how to make complicated molecules step by step. This process, called retrosynthesis, starts with the final molecule and works backward to find simpler building blocks. The authors created RetroGEF, a computer model that can predict how to break down a molecule into smaller parts by adding and changing atoms and bonds all at once. Unlike some older methods, RetroGEF doesn’t need a fixed size or order for these changes. Tests show RetroGEF can do this task better than previous methods.
Open → 2609.38484v1

RIDE improves drug molecule scaffold design preserving 3D shape

RIDE: Reference-Anchored Inference-Time Diffusion Editing for Scaffold Hopping

Abstract: Scaffold hopping is a critical task in drug discovery, which seeks to discover new, structurally distinct molecules that share key functional groups and similar 3D shape with a reference binding ligand. Existing diffusion-based scaffold hopping methods formulate the problem as conditional generation of scaffolds given the functional groups. However, they lack a principled mechanism to jointly enforce 2D structural novelty and preserve the 3D shape of the reference ligand. Here, we introduce RIDE, a Reference-anchored Inference-time Diffusion Editing framework for scaffold hopping. RIDE recovers the reference diffusion noise trajectory conditioned on the binding pocket and functional groups, selects an optimal trajectory segment for editing via noise perturbation, and conducts a value-guided scaffold sampling to generate new scaffolds. Extensive experimental results demonstrate that, compared to baselines, RIDE consistently generates scaffolds with lower 2D similarity and higher 3D similarity to the reference, with an average improvements of 11.7% and 7.3%, respectively. Further analysis reveals that RIDE can accommodate various reward functions, and can preserve 3D similarity even when this is not explicitly included in the reward. Two case studies illustrate RIDE's ability to generate distinct scaffolds with different structures and properties, and its ability to introduce substantial 2D variation while maintaining very high 3D similarity. RIDE is publicly available at https://anonymous.4open.science/r/RIDE-C8A0.

Mon 28 SeptArtificial IntelligenceMachine Learning
The gist
Designing new drug molecules often involves changing the core structure or scaffold while keeping important parts that interact with the body. The authors developed RIDE, a method that edits molecules by preserving their 3D shape but creating new 2D structures. This helps find distinct molecules that might work similarly but are structurally different, which is important for drug discovery. They show RIDE works better than other methods by producing more novel structures that still look alike in 3D.
Open → 2609.35623v1

TreeRef-BFN generates versatile 3D molecules from variable structures

TreeRef-BFN: Equivariance-Free De Novo Molecule Generation based on 2D Topology and Internal 3D Geometry

Abstract: De novo 3D molecular generation jointly models molecular size, topology, and geometry. Most methods pre-sample molecular size and generate Cartesian coordinates, limiting variable-size conditional tasks such as fragment completion and scaffold decoration while often relying on equivariant architectures. Internal-coordinate methods avoid rigid-body redundancy but typically require a known molecular graph or autoregressive construction, which may accumulate errors. We propose TreeRef, a tree-based molecular representation that assigns molecular topology and topology-dependent local 3D geometry to a naturally variable-size tree. RingRef nodes encode ring closures while preserving the tree structure, while Null nodes allow molecular size to emerge directly from node occupancy. Based on TreeRef, we develop TreeRef-BFN, a Bayesian Flow Network with a standard Transformer backbone that globally couples these locally defined variables and jointly generates discrete molecular variables and continuous local geometry. A single pretrained TreeRef-BFN supports unconditional generation and variable-size structure-conditioned 3D generation through masking alone, without retraining. Empirical studies demonstrate strong chemical validity, molecular stability, and diversity, accurate local geometric distributions, fast sampling, and competitive property-conditioned generation, establishing TreeRef-BFN as an efficient and flexible framework for 3D molecular generation.

Sat 26 SeptArtificial IntelligenceMachine Learning
The gist
Making new 3D molecules from scratch is hard because you have to decide their size, shape, and connections all at once. The authors created TreeRef-BFN, a method that represents molecules as trees capturing both their connections and small 3D details. This lets the method build molecules of different sizes and shapes more flexibly, without retraining for each task. It can quickly generate realistic and stable molecules with varied properties, useful for designing chemicals or drugs.
Open → 2609.32502v1

Geometric model predicts molecule shapes and sizes more accurately

Predicting Collision Cross Sections with GRACE: Geometric Residual Adduct Conditioning via Early-fusion

Abstract: Collision cross section (CCS), derived from ion mobility mass spectrometry, is a common descriptor for molecular annotation. Prediction is challenging for machine learning models because it reflects the size, shape, and ionization state of a gas-phase molecular ion. Most predictors either ignore explicit 3D structure or treat adduct identity as a late categorical feature, which limits their ability to capture adduct-dependent geometric effects. We present GRACE (Geometric Residual Adduct Conditioning via Early-fusion), a 3D CCS predictor that adapts a pretrained molecular geometry encoder using geometric residual adduct conditioning via early fusion. GRACE combines two inductive biases: a residual objective relative to an adduct-aware physical descriptor baseline and adduct conditioning within the encoder via a learned adduct token and low-rank attention adapters. We evaluate the model on a curated set of over 9,000 experimental molecule-adduct CCS records with random, scaffold, and adduct-sensitive splits designed to separate interpolation, scaffold generalization, and adduct-driven generalization. GRACE achieves the best mean percentage difference among the evaluated learned models on all three splits: 1.67% on the random split, 2.11% on the scaffold split, and 2.36% on the adduct-sensitive split. Diagnostic analyses suggest that residual learning stabilizes training by removing the dominant mass-CCS trend, while early fusion improves adduct-sensitive prediction relative to late fusion. Across four independent external test sets, GRACE shows consistently lower error than the other evaluated models. On a held-out set, GRACE also attains the lowest mean percent difference when compared with four previously reported physics-based workflows. These results support residual learning and encoder-level adduct conditioning as practical inductive biases for fast, accurate CCS prediction.

Thu 10 SeptMachine LearningArtificial Intelligence
The gist
Knowing how molecules travel through gas helps scientists identify them, but predicting this requires understanding their 3D shape and how they carry extra bits called adducts. The authors created a model called GRACE that learns from both the molecule's shape and details about these adducts early in its process, improving the predictions. Tested on thousands of examples, GRACE was more accurate than past models, especially when dealing with new or unusual molecules. This could help researchers analyze molecules faster and more reliably.
Open → 2609.12223v1