Reinforcing the Supply Chain of Umifenovir and Other Antiviral Drugs with Retrosynthetic Software

Abstract

The global disruption caused by the 2020 coronavirus pandemic stressed the supply chain of many products, including pharmaceuticals. Multiple drug repurposing studies for COVID-19 are now underway. If a winning therapeutic emerges, it is unlikely that the existing inventory of the medicine, or even the chemical raw materials needed to synthesize it, will be available in the quantities required. We used retrosynthetic software to arrive at alternate chemical supply chains for the antiviral drug umifenovir, as well as eleven other antiviral and anti-inflammatory drugs. We have experimentally validated four routes to umifenovir and one route to bromhexine. In several instances, the software utilizes C–H functionalization logic, and one route to umifenovir employs functionalization of six C–H bonds. The general strategy we apply can be used to identify distinct starting materials, and relieve stress on existing supply chains.

AI-Driven Synthetic Route Design Incorporated with Retrosynthesis Knowledge

“…These results clearly show that retrosynthesis knowledge effectively contributes to retrosynthetic analyses using ReTReK. However, the experimental validations 56 with targets that chemists are interested in and blind assessments 9 by trained chemists are not performed in this study; thus we plan to perform these validations to evaluate the performance of ReTReK on actual chemical synthesis in future work. Considering these evaluations and demonstrations, the ReTReK framework with integrated retrosynthesis knowledge has the potential to further improve the performance of datadriven CASP applications.…”

Rapid Planning and Analysis of High-Throughput Experiment Arrays for Reaction Discovery

“…In the example of Figure 3B, we optimized the penultimate step in our synthesis of umifenovir, 52 an oxidative indolization reaction between 14 and 15 to produce 16. Inspired by the conditions of Glorius, 61 a reaction array was performed using copper catalysts and ligand/additive combinations.…”

“…Reaction discovery and library synthesis campaigns utilizing standard 24-, 96-, 384and 1,536-well experiments are described in detail. [50][51][52][53][54][55]…”

Network search algorithms and scoring functions for advanced‐level computerized synthesis planning

“…Given the recent feats of AI approaches in many other areas, the outcome of the “retrosynthesis rivalry” has been somewhat curious and not in AI's favor. On one hand, the success of our “hybrid” expert‐MM‐QM‐ML approach has been documented by multiple experimental validations—by us 21,22,34 and others 35 —of complete, Chematica‐designed reaction routes leading not only to medicinally relevant targets 21,35 (Figure 1a) but also complex natural products, 22,34 at the forefront of modern organic synthesis (Figure 1b). Moreover, in, 22 we showed that world's leading experts were no longer able to distinguish the machine‐planned syntheses of very complex targets from those designed by humans.…”

“…Moreover, in, 22 we showed that world's leading experts were no longer able to distinguish the machine‐planned syntheses of very complex targets from those designed by humans. The program, rebranded as Synthia, is now commercialized and deployed by Merck KGaA in industry and academia 35,36 worldwide. On the other hand, the AI‐based efforts have been very prolific in proposing various NN architectures but the synthetic routes they produce continue to be limited to simple targets for which, one could argue, a competent chemist would not really need any computational support.…”