Predicting Retrosynthetic Pathways Using a Combined Linguistic Model and Hyper-Graph Exploration Strategy

Schwaller, Philippe; Petraglia, Riccardo; Zullo, Valerio; Nair, Vishnu H; Haeuselmann, Rico Andreas; Pisoni, Riccardo; Bekas, Costas; Iuliano, Anna; Laino, Teodoro

doi:10.26434/chemrxiv.9992489.v1

“…In other words, given a product molecule from the reaction database, the model correctly predicted the template that was extracted from the reaction that was recorded to produce the product molecule. This metric evaluates the ability of the model to "learn the dataset" on which it was trained, but, as has been discussed recently, 28 is not the only useful metric to evaluate retrosynthetic models. Figure 5 shows the top-k accuracy of baseline and pretrained models for the USPTO-Sm and USPTO-Lg datasets.…”

Section: Top-k Accuracymentioning

confidence: 99%

Data Augmentation and Pretraining for Template-Based Retrosynthetic Prediction in Computer-Aided Synthesis Planning

Fortunato

¹

,

Coley

²

,

Barnes

³

et al. 2020

J. Chem. Inf. Model.

View full text Add to dashboard Cite

This work presents efforts to augment the performance of data-driven machine learning algorithms for reaction template recommendation used in computer-aided synthesis planning software. Often, machine learning models designed to perform the task of prioritizing reaction templates or molecular transformations are focused on reporting high accuracy metrics for the one-to-one mapping of product molecules in reaction databases to the template extracted from the recorded reaction. The available templates that get selected for inclusion in these machine learning models have been previously limited to those that appear frequently in the reaction databases and exclude potentially useful transformations. By augmenting open-access datasets of organic reactions with artificially calculated template applicability and pretraining a template relevance neural network on this augmented applicability dataset, we report an increase in the template applicability recall and an increase in the diversity of predicted precursors. The augmentation and pretraining effectively teaches the neural network an increased set of templates that could theoretically lead to successful reactions for a given target. Even on a small dataset of well curated reactions, the data augmentation and pretraining methods resulted in an increase in top-1 accuracy, especially for rare templates, indicating these strategies can be very useful for small datasets. File list (2)download file view on ChemRxiv Data_Augmentation_and_Pre_Training_for_Template_... (563.68 KiB) download file view on ChemRxiv SI_Data_Augmentation_and_Pre_Training_for_Templat...

show abstract

“…Still, it indiscriminately regards other unobserved potentially feasible reactions as equally infeasible, resulting in a low recall. Essentially, retrosynthesis is a many-tomany problem (Thakkar et al 2022;Schwaller et al 2019b), where the target molecule M can potentially be synthesized through various distinct retro-strategies T , and vice versa. To mitigate the low recall issue, we aim to enhance the concept of template applicability by transforming the binary criteria of the observed ground truth P gt (M, T ) into a continuous approximation using a probabilistic model.…”

Section: Concept Enhancement For Label Shiftmentioning

confidence: 99%

RetroOOD: Understanding Out-of-Distribution Generalization in Retrosynthesis Prediction

Yu,

Yuan,

Wei

et al. 2024

AAAI

0

View full text Add to dashboard Cite

Machine learning-assisted retrosynthesis prediction models have been gaining widespread adoption, though their performances oftentimes degrade significantly when deployed in real-world applications embracing out-of-distribution (OOD) molecules or reactions. Despite steady progress on standard benchmarks, our understanding of existing retrosynthesis prediction models under the premise of distribution shifts remains stagnant. To this end, we first formally sort out two types of distribution shifts in retrosynthesis prediction and construct two groups of benchmark datasets. Next, through comprehensive experiments, we systematically compare state-of-the-art retrosynthesis prediction models on the two groups of benchmarks, revealing the limitations of previous in-distribution evaluation and re-examining the advantages of each model. More remarkably, we are motivated by the above empirical insights to propose two model-agnostic techniques that can improve the OOD generalization of arbitrary off-the-shelf retrosynthesis prediction algorithms. Our preliminary experiments show their high potential with an average performance improvement of 4.6%, and the established benchmarks serve as a foothold for further retrosynthesis prediction research towards OOD generalization.

show abstract

“…This metric evaluates the ability of the model to "learn the dataset" on which it was trained, but, as has been discussed recently, 28 is not the only useful metric to evaluate retrosynthetic models. Figure 6 shows the breakdown of top-100 accuracy for each data set by the original, unaugmented template popularity.…”

Section: Top-k Accuracymentioning

confidence: 99%

Data Augmentation and Pretraining for Template-Based Retrosynthetic Prediction in Computer-Aided Synthesis Planning

Fortunato

¹

,

Coley

²

,

Barnes

³

et al. 2020

Preprint

View full text Add to dashboard Cite

This work presents efforts to augment the performance of data-driven machine learning algorithms for reaction template recommendation used in computer-aided synthesis planning software. Often, machine learning models designed to perform the task of prioritizing reaction templates or molecular transformations are focused on reporting high accuracy metrics for the one-to-one mapping of product molecules in reaction databases to the template extracted from the recorded reaction. The available templates that get selected for inclusion in these machine learning models have been previously limited to those that appear frequently in the reaction databases and exclude potentially useful transformations. By augmenting open-access datasets of organic reactions with artificially calculated template applicability and pretraining a template relevance neural network on this augmented applicability dataset, we report an increase in the template applicability recall and an increase in the diversity of predicted precursors. The augmentation and pretraining effectively teaches the neural network an increased set of templates that could theoretically lead to successful reactions for a given target. Even on a small dataset of well curated reactions, the data augmentation and pretraining methods resulted in an increase in top-1 accuracy, especially for rare templates, indicating these strategies can be very useful for small datasets.

show abstract

Predicting Retrosynthetic Pathways Using a Combined Linguistic Model and Hyper-Graph Exploration Strategy

Cited by 7 publications

References 64 publications

Data Augmentation and Pretraining for Template-Based Retrosynthetic Prediction in Computer-Aided Synthesis Planning

Data Augmentation and Pretraining for Template-Based Retrosynthetic Prediction in Computer-Aided Synthesis Planning

RetroOOD: Understanding Out-of-Distribution Generalization in Retrosynthesis Prediction

Data Augmentation and Pretraining for Template-Based Retrosynthetic Prediction in Computer-Aided Synthesis Planning

Contact Info

Product

Resources

About