More Like this

1. ClassicalGSG : Prediction of log P using classical molecular force fields and geometric scattering for graphs

   https://doi.org/10.1002/jcc.26519  

   Donyapour, Nazanin; Hirn, Matthew; Dickson, Alex (March 2021, Journal of Computational Chemistry)

   Abstract This work examines methods for predicting the partition coefficient (logP) for a dataset of small molecules. Here, we use atomic attributes such as radius and partial charge, which are typically used as force field parameters in classical molecular dynamics simulations. These atomic attributes are transformed into index‐invariant molecular features using a recently developed method called geometric scattering for graphs (GSG). We call this approach “ClassicalGSG” and examine its performance under a broad range of conditions and hyperparameters. We train ClassicalGSG logPpredictors with neural networks using 10,722 molecules from the OpenChem dataset and apply them to predict the logPvalues from four independent test sets. The ClassicalGSG method's performance is compared to a baseline model that employs graph convolutional networks. Our results show that the best prediction accuracies are obtained using atomic attributes generated with the CHARMM generalized force field and 2D molecular structures. 

   more » « less
2. A graph based approach to model charge transport in semiconducting polymers

   https://doi.org/10.1038/s41524-022-00714-w  

   Noruzi, Ramin; Lim, Eunhee; Pokuri, Balaji Sesha Sarath; Chabinyc, Michael L.; Ganapathysubramanian, Baskar (March 2022, npj Computational Materials)

   Abstract Charge transport in molecular solids, such as semiconducting polymers, is strongly affected by packing and structural order over several length scales. Conventional approaches to modeling these phenomena range from analytical models to numerical models using quantum mechanical calculations. While analytical approaches cannot account for detailed structural effects, numerical models are expensive for exhaustive (and statistically significant) analysis. Here, we report a computationally scalable methodology using graph theory to explore the influence of molecular ordering on charge mobility. This model accurately reproduces the analytical results for transport in nematic and isotropic systems, as well as experimental results of the dependence of the charge carrier mobility on orientation correlation length for polymers. We further model how defect distribution (correlated and uncorrelated) in semiconducting polymers can modify the mobility, predicting a critical defect density above which the mobility plummets. This work enables rapid (and computationally extensible) evaluation of charge mobility semiconducting polymer devices. 

   more » « less
3. Super‐ and Ferroelastic Organic Semiconductors for Ultraflexible Single‐Crystal Electronics

   https://doi.org/10.1002/anie.202004083  

   Park, Sang Kyu; Sun, Hong; Chung, Hyunjoong; Patel, Bijal B.; Zhang, Fengjiao; Davies, Daniel W.; Woods, Toby J.; Zhao, Kejie; Diao, Ying (May 2020, Angewandte Chemie International Edition)

   Abstract Like silicon, single crystals of organic semiconductors are pursued to attain intrinsic charge transport properties. However, they are intolerant to mechanical deformation, impeding their application in flexible electronic devices. Such contradictory properties, namely exceptional molecular ordering and mechanical flexibility, are unified in this work. We found that bis(triisopropylsilylethynyl)pentacene (TIPS‐P) crystals can undergo mechanically induced structural transitions to exhibit superelasticity and ferroelasticity. These properties arise from cooperative and correlated molecular displacements and rotations in response to mechanical stress. By utilizing a bending‐induced ferroelastic transition of TIPS‐P, flexible single‐crystal electronic devices were obtained that can tolerate strains (ϵ) of more than 13 % while maintaining the charge carrier mobility of unstrained crystals (μ>0.7 μ₀). Our work will pave the way for high‐performance ultraflexible single‐crystal organic electronics for sensors, memories, and robotic applications. 

   more » « less
4. Intramolecular Charge Transfer in Antiaromatic Donor/Acceptor‐Fused s ‐Indacenes

   https://doi.org/10.1002/anie.202420989  

   Demachkie, Isabella_S; Miller, Michael_P; Warren, Gabrielle_I; Barker, Joshua_E; Strand, Eric_T; Zakharov, Lev_N; Haley, Michael_M (December 2024, Angewandte Chemie International Edition)

   Abstract Herein we report the synthesis and characterization of four donor/acceptor‐fuseds‐indacenes via the late‐stage oxidation of a family of unsymmetrical benzofuran/benzothiophene‐s‐indacene regioisomers. A thorough study of their properties through experimental and computational analysis has revealed the effect of asymmetry on the molecular properties associated with antiaromaticity, as well as a strong correlation between antiaromaticity and intramolecular charge transfer (ICT). The strength of the charge transfer depends on the fusion orientation of the donor and acceptor motifs relative to thes‐indacene core. The two most antiaromatic oxidized isomers exhibit strong evidence of ICT with 30 and 40 nm solvatochromic shifts. 

   more » « less
5. Rationalizing Graph Neural Networks with Data Augmentation

   https://doi.org/10.1145/3638781  

   Liu, Gang; Inae, Eric; Luo, Tengfei; Jiang, Meng (May 2024, ACM Transactions on Knowledge Discovery from Data)

   Graph rationales are representative subgraph structures that best explain and support the graph neural network (GNN) predictions. Graph rationalization involves the joint identification of these subgraphs during GNN training, resulting in improved interpretability and generalization. GNN is widely used for node-level tasks such as paper classification and graph-level tasks such as molecular property prediction. However, on both levels, little attention has been given to GNN rationalization and the lack of training examples makes it difficult to identify the optimal graph rationales. In this work, we address the problem by proposing a unified data augmentation framework with two novel operations on environment subgraphs to rationalize GNN prediction. We define the environment subgraph as the remaining subgraph after rationale identification and separation. The framework efficiently performs rationale–environment separation in therepresentation spacefor a node’s neighborhood graph or a graph’s complete structure to avoid the high complexity of explicit graph decoding and encoding. We conduct experiments on 17 datasets spanning node classification, graph classification, and graph regression. Results demonstrate that our framework is effective and efficient in rationalizing and enhancing GNNs for different levels of tasks on graphs. 

   more » « less