The results establish Molexar as a practical unified foundation for computational chemistry and drug-design workflows and establish Molexar as a practical unified foundation for computational chemistry and drug-design workflows.
Abstract
Molecular generation is a central challenge in drug discovery, requiring models that explore vast chemical space while satisfying diverse design constraints. We present Molexar, a unified multimodal molecular foundation model built on Fragment-SELFIES, a robust, fragment-aware molecular language with validity-preserving decoding and explicit fragment structure. A pretrained autoregressive decoder learns the Fragment-SELFIES syntax and molecular distribution; supervised fine-tuning (SFT) then trains the same decoder on condition-molecule pairs spanning scalar molecular properties, pharmacophore fingerprints, protein sequences, and binding pockets, injecting each condition by in-place replacement of value-token embeddings so that all generation modes share one autoregressive path. Molexar achieves strong efficiency at a small parameter count while matching or exceeding larger models. The pretrained model reaches 100% validity and high drug-likeness in unconditional and fragment-constrained generation; the SFT model follows single- and multi-property instructions and remains competitive on target-conditioned generation on the CrossDocked2020 test set. On MolGenBench, Molexar further generates molecules with favorable safety and potency. These results establish Molexar as a practical unified foundation for computational chemistry and drug-design workflows.
De novo drug design aims to generate molecules targeting specific protein pockets while retaining chemical plausibility and drug-like properties. Recent 3D structure-based generative methods explicitly model pocket-ligand geometry, but this does not always translate into chemically realistic or practically usable candidate molecules. Molecular language models provide a complementary sequence-based alternative. However, it remains unclear whether sequence-based pocket information can effectively guide ligand generation, whether multistage training improves pocket-specific generation, and whether docking-guided reinforcement learning can be integrated into a practical generation pipeline. We introduce PockLigGPT, a GPT-based framework for pocket-sequence-conditioned molecular generation. Rather than producing fixed 3D coordinates, PockLigGPT formulates ligand design as a sequence-generation problem conditioned on the amino acid composition of the protein pocket. The model is trained in four stages: large-scale chemical pretraining based on ZINC20; bioactivity-oriented adaptation based on ChEMBL; pocket-sequence-conditioned fine-tuning using binding-pocket amino acid sequences paired with ligands; and, finally, pocket-specific docking-guided reinforcement learning using AutoDock Vina-based rewards. PockLigGPT achieves competitive docking-oriented performance under a standardized evaluation protocol while maintaining chemical plausibility, favorable physicochemical profiles, and Lipinski-based drug-likeness. Docking studies on Alzheimer’s disease-associated targets and token-level analyses further support the utility of PockLigGPT for de novo drug design.
Pablo Varas Pardo, Guillermo Marcos-Ayuso, Eugenia Ulzurrun et al.· Journal of Chemical Informat...· 0 citations
Generative molecular design is shaped by simple proxy benchmarks for drug-like properties and models pretrained on large pharmaceutical datasets. This combination yields strong benchmark metrics but limits transferability to domains structurally distinct from drug discovery. To overcome this limitation and drive discovery toward real, scientifically grounded targets, we introduce the Nanotechnology Molecular Optimization (NMO) Benchmark, which bridges machine learning (ML) and quantum materials science. NMO acts simultaneously as a rigorous testbed for the ML community and a discovery engine for nanotechnology research. The suite replaces proxy oracles with quantum simulations and introduces strict protocols that prioritize scientific utility over leaderboard-oriented overfitting. The physics-based NMO tasks impose hard structural constraints and rugged fitness landscapes, posing fundamentally new requirements on generative models. Notably, advanced molecular optimization methods underperform much simpler approaches on the NMO tasks. We develop a new baseline method identifying the critical components to solve the NMO tasks, including a novel representation for modeling structural constraints and a domain-agnostic pretraining strategy to eliminate pharmaceutical dataset bias. Our results surpass state-of-the-art physical properties and reveal previously unknown structural motifs, offering new insights for the nanotechnology community and demonstrating that ML can drive genuine scientific discovery.
Matthias Blaschke, Daniel Kienzle, Zsuzsanna Koczor-Benda et al.· arXiv.org· 0 citations
Accurate molecular property prediction is a cornerstone of modern chemical science, driving progress in drug discovery, materials design, and environmental research. Yet, most existing models remain unimodal, while multimodal approaches often rely on simple aggregation, leaving much of the complementary chemical information underexploited. In this work, we present a multimodal feature fusion framework that unites the strengths of deep chemical language processing (CLP) models and molecular fingerprints, integrating sequential and structural representations for more comprehensive molecular characterization. Unlike previous heuristic combinations, our framework systematically investigates the principles of effective cross-modal fusion. We benchmark ten CLP architectures and eight fingerprint types through exhaustive combinatorial search to identify the most synergistic configurations. This exploration shows that aggregating multiple models does not necessarily improve performance; instead, successful fusion requires data-aware design guided by feature integration and complementarity. The proposed strategy effectively couples sequential features learned from SMILES with structural information captured by molecular fingerprints, resulting in a coherent and chemically interpretable molecular representation. Evaluated across 60 datasets from MoleculeNet and TOXRIC, our fusion models deliver consistent and substantial gains over state-of-the-art baselines. Beyond outperforming existing architectures, this work provides conceptual insights and practical guidelines for multimodal fusion in molecular property prediction, highlighting the importance of efficient fusion strategies in building robust and generalizable molecular models.Scientific contributionThis study provides a large-scale empirical evaluation of multimodal feature fusion for molecular property classification by systematically integrating SMILES-based chemical language representations with fingerprint-based structural descriptors across 60 benchmark datasets. The framework introduces a data-aware combinatorial fusion strategy to identify task-specific complementary feature combinations, improving robustness and interpretability compared with unimodal models and baseline approaches. The results clarify how sequence-based and structure-based molecular representations complement each other, providing practical guidance for designing multimodal models in cheminformatics.
Jing Liu, Li Xue, Yin Wang et al.· Journal of Cheminformatics· 0 citations
Macrocyclic peptides are an increasingly important therapeutic modality, but existing computational methods for modeling their structures and properties are limited in scope and do not generalize well across the synthetically accessible chemical space. In this work, we introduce Vilya-1, a deep learning model that addresses two central challenges in macrocycle design: sampling biologically relevant conformations across arbitrary chemistries and predicting key developability properties such as membrane permeability. Vilya-1 operates on a uniform all-atom representation and is trained on heterogeneous structural datasets spanning diverse topologies and chemical classes. Across a broad set of macrocycles composed of canonical and non-canonical residues, Vilya-1 substantially improves geometric accuracy relative to physics-based methods, co-folding networks, and deep-learning conformer generators, while maintaining broad chemical coverage that extends to small molecules. Vilya-1 also supports generative applications, enabling the design of novel macrocycles with tailored chemical, structural, and property profiles. Together, these capabilities establish Vilya-1 as a foundation model for accelerating the development of next-generation macrocycle therapeutics.
Vilya Research Pascal Sturmfels, M. Salem, Naozumi Hiranuma et al.· 1 citation
Structure-based generative models (SBGMs) hold great promises for accelerating drug discovery by enabling target-aware molecular design. However, existing approaches face fundamental challenges: three-dimensional graph-based models can explicitly incorporate protein structural information but often generate chemically implausible molecules due to limited training data, while chemical language models (CLMs) produce chemically plausible molecules but struggle to effectively leverage three-dimensional structural information for structure-conditioned generation and hard to incorporate lead optimization functionality due to the nature of SMILES string. Here, we present StructureSAFE, a structure-aware chemical language model that resolves this trade-off by integrating protein structural and evolutionary encoders with the SAFE molecular representation via pretraining and finetuning training scheme, enabling both de novo hit identification and a comprehensive suite of lead optimization subtasks within a unified framework. Comprehensive benchmarking on the MolGenBench dataset demonstrates that StructureSAFE achieves state-of-the-art (SOTA) performance across multiple metrics, with particularly pronounced improvements in chemical plausibility relative to graph-based models lacking pretraining. Evaluation on a rigorously constructed held-out test set further confirms its ability to generate drug-like, synthetically accessible molecules with competitive predicted binding affinities for previously unseen targets on both hit identification and lead optimization setting. In silico case studies across four therapeutically relevant targets validate its capacity to generate chemically plausible molecules that recapitulate key binding interactions of known high-affinity ligands while proposing novel interactions for potential better affinity and exploring previously unknown regions of chemical space. Taking together, StructureSAFE represents a versatile and practical tool to provide high-quality candidate molecules for augmenting medicinal chemistry workflows in both hit identification and lead optimization campaigns.
Bo Yang, Ke Xu, Chijian Xiang et al.· bioRxiv· 0 citations
This review elucidates the paradigm shift in drug discovery from serendipitous exploration to rational, structure-driven polypharmacological molecular engineering, thereby providing a clear, structured guide for navigating the complexities of next-generation therapeutics.
Tianming Han, Zhijie Pan, Wenchi Ge et al.· 0 citations