Strategy for Efficient Discovery of Cocrystals via a Network-Based Recommendation Model

Abstract

Experimental screening of cocrystals is usually laborious and time-consuming; therefore, it is urgent to develop effective in silico predictive models to guide cocrystal discovery. In this study, network-based recommendation models were proposed to predict new cocrystals for molecules in cocrystal network. The local random walk (LRW) recommender algorithm was first confirmed as an effective model in cocrystal design. The algorithmic principle of LRW could capture the supramolecular synthon mechanisms in the cocrystal system and grasp the structural features of the cocrystal network, thus possessing satisfactory predictive capability. Various pharmaceutical cocrystals reported in the recent literature could be distinguished by our model, which demonstrates the good generalization capability inherent in our approach. As a case study, new cocrystals for apatinib were predicted and subsequently obtained. The consistency between prediction and experimental results highlighted the accuracy and practicability of the predictive model. Particularly, our predictive model is competitive in computational time and easy to implement. In summary, our network-based recommendation model would be an effective tool to guide experimental cocrystal screening and improve the efficiency of cocrystal discovery.

SMINBR: An Integrated Network and Chemoinformatics Tool Specialized for Prediction of Two-Component Crystal Formation

“…According to this criterion, the predictive capability of SMINBR was further compared with HBP, HBC, HBE, and LRW by conducting diclofenac and nevirapine cocrystal prediction tasks. 28,37 The detailed predictive results are provided in Tables S7 and S8 in the Supporting Information, and these results are summarized in Table 3.…”

“…Top 10 interaction partners for APA were recommended (Table 4), in which fumaric acid, succinic acid, adipic acid, oxalic acid, benzoic acid, malonic acid, glutaric acid, 4-aminobenzoic acid, and 4-hydroxybenzoic acid have been found to form cocrystals with APA in previous studies. 28,42 In this study, paradioxybenzene (PDB) was subject to experimental screening and found to form a new cocrystal with APA. This new APA-PDB cocrystal was analyzed by powder X-ray diffraction (PXRD) to distinguish it from the parent phases (see Figure 6).…”

“…According to our previous study, molecular pairs are easy to form intermolecular interactions when they possess complementary interaction interfaces instead of similar structural features. 28 As reflected in the network, these connected molecular pairs are often linked by three-length paths (V and Y in Figure 1E).…”

“…In our previous work, a network-based recommendation model based on local random walk (LRW) was used to guide experimental cocrystal screening. 28 These network-based models with low computational complexity and satisfactory predictive capability have been proven to be useful alternatives to the existing methods. However, there are several intrinsic pitfalls in such traditional social network-derived models.…”

Pharmaceutical cocrystals: A review of preparations, physicochemical properties and applications

“…However, the selection of coforms has still largely relied on trial and error, which is time consuming and labor-intensive. Recently, a variety of computer-assisted approaches have emerging as attractive tools to accelerate the screening process of cocrystals 214 , 215 , 216 , 217 , 218 . For example, the artificial neural network models are developed to predict cocrystal formation by analyzing a network of coformers extracted from the Cambridge Structural Database (CSD) 218 .…”