Lu Hao, Wei Zhiqiang, Zhang Kun, Wang Xuze, Ali Liaqat, Liu Hao
IEEE/ACM Trans Comput Biol Bioinform. 2024 Nov-Dec;21(6):2235-2245. doi: 10.1109/TCBB.2024.3455381. Epub 2024 Dec 10.
Retrosynthesis prediction is a fundamental problem in organic chemistry and drug synthesis. We proposed an end-to-end deep learning model called CTsynther (Contrastive Transformer for single-step retrosynthesis prediction model) that could provide single-step retrosynthesis prediction without external reaction templates or specialized knowledge. The model introduced the concept of contrastive learning in Transformer architecture and employed a contrastive learning language representation model at the SMILES sentence level to enhance model inference by learning similarities and differences between various samples. Mixed global and local attention mechanisms allow the model to capture features and dependencies between different atoms to improve generalization. We further investigated the embedding representations of SMILES learned automatically from the model. Visualization results show that the model could effectively acquire information about identical molecules and improve prediction performance. Experiments showed that the accuracy of retrosynthesis reached 53.5% and 64.4% for with and without reaction types, respectively. The validity of the predicted reactants is improved, showing competitiveness compared with semi-template methods.
逆合成预测是有机化学和药物合成中的一个基本问题。我们提出了一种名为CTsynther（用于单步逆合成预测模型的对比Transformer）的端到端深度学习模型，该模型无需外部反应模板或专业知识即可提供单步逆合成预测。该模型在Transformer架构中引入了对比学习的概念，并在SMILES句子级别采用了对比学习语言表示模型，通过学习各种样本之间的异同来增强模型推理。混合全局和局部注意力机制使模型能够捕获不同原子之间的特征和依赖性，从而提高泛化能力。我们进一步研究了从模型中自动学习到的SMILES嵌入表示。可视化结果表明，该模型能够有效地获取相同分子的信息并提高预测性能。实验表明，有反应类型和无反应类型时逆合成的准确率分别达到53.5%和64.4%。预测反应物的有效性得到了提高，与半模板方法相比具有竞争力。
