Prediction of Compound Synthesis Accessibility Based on Reaction Knowledge Graph.

With the increasing application of deep-learning-based generative models for de novo molecule design, the quantitative estimation of molecular synthetic accessibility (SA) has become a crucial factor for prioritizing the structures generated from generative models. It is also useful for helping in the prioritization of hit/lead compounds and guiding retrosynthesis analysis. In this study, based on the USPTO and Pistachio reaction datasets, a chemical reaction network was constructed for the identification of the shortest reaction paths (SRP) needed to synthesize compounds, and different SRP cut-offs were then used as the threshold to distinguish a organic compound as either an easy-to-synthesize (ES) or hard-to-synthesize (HS) class. Two synthesis accessibility models (DNN-ECFP model and graph-based CMPNN model) were built using deep learning/machine learning algorithms. Compared to other existing synthesis accessibility scoring schemes, such as SYBA, SCScore, and SAScore, our results show that CMPNN (ROC AUC: 0.791) performs better than SYBA (ROC AUC: 0.76), albeit marginally, and outperforms SAScore and SCScore. Our prediction models based on historical reaction knowledge could be a potential tool for estimating molecule SA.

Synthesis-accessibility-oriented design of c-Jun N-terminal kinase 1 inhibitor.

Idiopathic pulmonary fibrosis (IPF) is a severe and progressive lung disease with poor prognosis and limited treatment options. The c-Jun N-Terminal Kinase 1 (JNK1), a key component of the MAPK pathway, has been implicated in the pathogenesis of IPF and represents a potential therapeutic target. However, the development of JNK1 inhibitors has been slowed, partly due to synthetic complexity in medicinal chemistry modification. Here, we report a synthesis-accessibility-oriented strategy for designing JNK1 inhibitors based on computational prediction of synthetic feasibility and fragment-based molecule generation. This strategy led to the discovery of several potent JNK1 inhibitors, such as compound C6 (IC50 = 33.5 nM), which exhibited comparable activity to the clinical candidate CC-90001 (IC50 = 24.4 nM). The anti-fibrotic effect of C6 was further confirmed in animal model of pulmonary fibrosis. Moreover, compound C6 could be synthesized in only two steps, compared to nine steps for CC-90001. Our findings suggest that compound C6 is a promising lead for further optimization and development as a novel anti-fibrotic agent targeting JNK1. In addition, the discovery of C6 also demonstrates the feasibility of synthesis-accessibility-oriented strategy in lead discovery.