Emma Granqvist, Rocío Mercado and Samuel Genheden
Digital Discovery, 2026, Advance Article
Abstract
We present RetroSynFormer, a novel approach to multi-step retrosynthesis planning. Here, we express the task of iteratively breaking down a compound into building blocks as a sequence-modeling problem and train a Decision Transformer-based model.
Jason D. Shields, Rachel Howells, Gillian Lamont, Yin Leilei, Andrew Madin, Christopher E. Reimann, Hadi Rezaei, Tristan Reuillon, Bryony Smith, Clare Thomson, Yuting Zheng and Robert E. Ziegler
RSC Med. Chem., 2024,15, 1085-1095
From themed collection:
Highlights from Industry
Abstract
The AI retrosynthesis tool AiZynth has made positive impacts on AstraZeneca drug discovery projects. This opinion provides some examples and discusses how AI retrosynthesis fits into pharmaceutical research.
Annie M. Westerlund, Lakshidaa Saigiridharan and Samuel Genheden
Chem. Sci., 2025,16, 14655-14667
Abstract
This work presents a novel strategy for human-guided multistep retrosynthesis via prompting, using a disconnection-aware transformer and multi-objective Monte Carlo Tree Search.
Samuel Genheden and Jason D. Shields
Digital Discovery, 2025,4, 46-53
Abstract
A simple method to calculate a similarity metric between two synthetic routes, based on bonds formed and grouping of atoms in the target compound.
Gökçe Geylan, Mikhail Kabeshov, Samuel Genheden, Christos Kannas, Thierry Kogej, Leonardo De Maria, Florian David and Ola Engkvist
Chem. Sci., 2025,16, 17927-17938
Abstract
NNAA-Synth is a synthesis assistance tool for SPPS-compatible non-natural amino acids that integrates protection strategies and evaluates reaction feasibility.