Loading
Life sciences · Preprint
arXiv · August 17, 2026
Early or partial results. Treat as a signal, not a conclusion.
RetroMPA is a model-agnostic computational filter that improves retrosynthesis prediction accuracy by 5.50% on USPTO-50K and 2.03% on USPTO-Full by integrating chemical knowledge into existing deep learning models. Preliminary wet-lab experiments demonstrate feasibility for three classic reaction types but lack systematic design, control, or statistical validation.
Computational framework development with post-hoc validation on benchmark datasets and uncontrolled wet-lab synthesis experiments. Computational: eight representative retrosynthesis models. Experimental: three classic organic reaction types (Suzuki-Miyaura coupling, Bucherer reaction, Friedel-Crafts acylation); substrate-level details and n not specified.. Intervention: RetroMPA: a model-agnostic, property-aware auxiliary filter module that recalibrates predictions from existing retrosynthesis models by leveraging a property-aware latent embedding space.. Compared with: Unmodified outputs from the same eight retrosynthesis models without RetroMPA enhancement..
Top-1 accuracy improvement of average 5.50% across eight retrosynthesis models on USPTO-50K Average improvement of about 2.03% on large-scale USPTO-Full dataset across template-based and template-free architectures Wet-lab experiments confirmed viable substrate combinations for Suzuki-Miyaura coupling, Bucherer reaction, and Friedel-Crafts acylation
Safety was not reported in the material analysed. Check the source before drawing any conclusion about harm.
The source did not state who this applies to in practice.
A computational framework with algorithmic improvements on benchmark datasets and limited wet-lab validation, lacking peer review and clinical/translational endpoints.
As stated by the source record.
Quoted from the source exactly as published.
Graded across the dimensions that decide whether you should act, each from what the source actually supports. There is no single score, and where a dimension was not assessed it says so.
Retrosynthesis is a cornerstone of drug discovery and organic synthesis. While data-driven deep learning models have shown remarkable progress, they autonomously learn reaction patterns from extensive datasets with limited integration of established chemical knowledge as priors. To address this limitation, we introduce RetroMPA, a molecular property-aware, post-hoc enhancement module that injects chemical knowledge into the retrosynthesis pipeline. Rather than functioning as an independent SMILES sequence generator, RetroMPA is a broadly applicable, model-agnostic chemical filter designed to recalibrate and optimize the predictive pathways of existing algorithms. This plug-and-play framework integrates seamlessly with a range of data-driven retrosynthesis methods, enhancing outputs without modifying model architecture or requiring resource-intensive retraining. By leveraging a property-aware latent embedding space, RetroMPA consistently improves top-1 accuracy across eight representative retrosynthesis models by an average of 5.50% on USPTO-50K. Furthermore, we validate its scalability on the large-scale USPTO-Full dataset, achieving an average improvement of about 2.03% across both template-based and template-free architectures. Wet-lab experiments provide preliminary support for the practical utility of the framework. These syntheses confirmed viable, previously unreported substrate combinations for classic reaction paradigms---specifically, Suzuki-Miyaura coupling, Bucherer reaction, and Friedel-Crafts acylation---suggesting that RetroMPA can operate beyond mere data fitting. The code is open-sourced at https://github.com/MengzhouLu/RetroMPA.
Taken from the source record, never inferred. Follow any of these and new work involving them reaches your briefing.