PD-1-Expressing SARS-CoV-2-Specific CD8+ T Cells Are Not Exhausted, but Functional in Patients with COVID-19.

Memory T cell responses have been demonstrated in COVID-19 convalescents, but ex vivo phenotypes of SARS-CoV-2-specific T cells have been unclear. We detected SARS-CoV-2-specific CD8+ T cells by MHC class I multimer staining and examined their phenotypes and functions in acute and convalescent COVID-19. Multimer+ cells exhibited early differentiated effector-memory phenotypes in the early convalescent phase. The frequency of stem-like memory cells was increased among multimer+ cells in the late convalescent phase. Cytokine secretion assays combined with MHC class I multimer staining revealed that the proportion of interferon-γ (IFN-γ)-producing cells was significantly lower among SARS-CoV-2-specific CD8+ T cells than those specific to influenza A virus. Importantly, the proportion of IFN-γ-producing cells was higher in PD-1+ cells than PD-1- cells among multimer+ cells, indicating that PD-1-expressing, SARS-CoV-2-specific CD8+ T cells are not exhausted, but functional. Our current findings provide information for understanding of SARS-CoV-2-specific CD8+ T cells elicited by infection or vaccination.

A LaCl3-based lithium superionic conductor compatible with lithium metal.

Inorganic superionic conductors possess high ionic conductivity and excellent thermal stability but their poor interfacial compatibility with lithium metal electrodes precludes application in all-solid-state lithium metal batteries1,2. Here we report a LaCl3-based lithium superionic conductor possessing excellent interfacial compatibility with lithium metal electrodes. In contrast to a Li3MCl6 (M = Y, In, Sc and Ho) electrolyte lattice3-6, the UCl3-type LaCl3 lattice has large, one-dimensional channels for rapid Li+ conduction, interconnected by La vacancies via Ta doping and resulting in a three-dimensional Li+ migration network. The optimized Li0.388Ta0.238La0.475Cl3 electrolyte exhibits Li+ conductivity of 3.02 mS cm-1 at 30 °C and a low activation energy of 0.197 eV. It also generates a gradient interfacial passivation layer to stabilize the Li metal electrode for long-term cycling of a Li-Li symmetric cell (1 mAh cm-2) for more than 5,000 h. When directly coupled with an uncoated LiNi0.5Co0.2Mn0.3O2 cathode and bare Li metal anode, the Li0.388Ta0.238La0.475Cl3 electrolyte enables a solid battery to run for more than 100 cycles with a cutoff voltage of 4.35 V and areal capacity of more than 1 mAh cm-2. We also demonstrate rapid Li+ conduction in lanthanide metal chlorides (LnCl3; Ln = La, Ce, Nd, Sm and Gd), suggesting that the LnCl3 solid electrolyte system could provide further developments in conductivity and utility.

The Raf-like MAPKKKs STY8, STY17, and STY46 negatively regulate Botrytis cinerea resistance by limiting MKK7 protein accumulation in Arabidopsis.

Plant diseases, which seriously damage crop production, are in most cases caused by fungal pathogens. In this study, we found that the Raf-like MAPKKKs STY8 (SERINE/THREONINE/TYROSINE KINASE 8), STY17, and STY46 negatively regulate resistance to the fungal pathogen Botrytis cinerea through jasmonate response in Arabidopsis. Moreover, STY8/STY17/STY46 homologs negatively contribute to chitin signaling. We further identified MKK7 as the MAPKK component interacting with STY8/STY17/STY46 homologs. MKK7 positively contributes to resistance to B. cinerea and chitin signaling. Furthermore, we found that STY8/STY17/STY46 homologs negatively affect the accumulation of MKK7, in accordance with the opposite roles of MKK7 and STY8/STY17/STY46 homologs in defense against B. cinerea. These results provide new insights into the mechanisms precisely regulating plant immunity via Raf-like MAPKKKs.

High-accuracy protein model quality assessment using attention graph neural networks.

Great improvement has been brought to protein tertiary structure prediction through deep learning. It is important but very challenging to accurately rank and score decoy structures predicted by different models. CASP14 results show that existing quality assessment (QA) approaches lag behind the development of protein structure prediction methods, where almost all existing QA models degrade in accuracy when the target is a decoy of high quality. How to give an accurate assessment to high-accuracy decoys is particularly useful with the available of accurate structure prediction methods. Here we propose a fast and effective single-model QA method, QATEN, which can evaluate decoys only by their topological characteristics and atomic types. Our model uses graph neural networks and attention mechanisms to evaluate global and amino acid level scores, and uses specific loss functions to constrain the network to focus more on high-precision decoys and protein domains. On the CASP14 evaluation decoys, QATEN performs better than other QA models under all correlation coefficients when targeting average LDDT. QATEN shows promising performance when considering only high-accuracy decoys. Compared to the embedded evaluation modules of predicted ${C}_{\alpha^{-}} RMSD$ (pRMSD) in RosettaFold and predicted LDDT (pLDDT) in AlphaFold2, QATEN is complementary and capable of achieving better evaluation on some decoy structures generated by AlphaFold2 and RosettaFold. These results suggest that the new QATEN approach can be used as a reliable independent assessment algorithm for high-accuracy protein structure decoys.

Leveraging scaffold information to predict protein-ligand binding affinity with an empirical graph neural network.

Protein-ligand binding affinity prediction is an important task in structural bioinformatics for drug discovery and design. Although various scoring functions (SFs) have been proposed, it remains challenging to accurately evaluate the binding affinity of a protein-ligand complex with the known bound structure because of the potential preference of scoring system. In recent years, deep learning (DL) techniques have been applied to SFs without sophisticated feature engineering. Nevertheless, existing methods cannot model the differential contribution of atoms in various regions of proteins, and the relationship between atom properties and intermolecular distance is also not fully explored. We propose a novel empirical graph neural network for accurate protein-ligand binding affinity prediction (EGNA). Graphs of protein, ligand and their interactions are constructed based on different regions of each bound complex. Proteins and ligands are effectively represented by graph convolutional layers, enabling the EGNA to capture interaction patterns precisely by simulating empirical SFs. The contributions of different factors on binding affinity can thus be transparently investigated. EGNA is compared with the state-of-the-art machine learning-based SFs on two widely used benchmark data sets. The results demonstrate the superiority of EGNA and its good generalization capability.

MINDG: a drug-target interaction prediction method based on an integrated learning algorithm.

MOTIVATION: Drug-target interaction (DTI) prediction refers to the prediction of whether a given drug molecule will bind to a specific target and thus exert a targeted therapeutic effect. Although intelligent computational approaches for drug target prediction have received much attention and made many advances, they are still a challenging task that requires further research. The main challenges are manifested as follows: (i) most graph neural network-based methods only consider the information of the first-order neighboring nodes (drug and target) in the graph, without learning deeper and richer structural features from the higher-order neighboring nodes. (ii) Existing methods do not consider both the sequence and structural features of drugs and targets, and each method is independent of each other, and cannot combine the advantages of sequence and structural features to improve the interactive learning effect. RESULTS: To address the above challenges, a Multi-view Integrated learning Network that integrates Deep learning and Graph Learning (MINDG) is proposed in this study, which consists of the following parts: (i) a mixed deep network is used to extract sequence features of drugs and targets, (ii) a higher-order graph attention convolutional network is proposed to better extract and capture structural features, and (iii) a multi-view adaptive integrated decision module is used to improve and complement the initial prediction results of the above two networks to enhance the prediction performance. We evaluate MINDG on two dataset and show it improved DTI prediction performance compared to state-of-the-art baselines. AVAILABILITY AND IMPLEMENTATION: https://github.com/jnuaipr/MINDG.

Dimensional and Doping Engineering of Chiral Perovskites with Enhanced Spin Selectivity for Green Emissive Spin Light-Emitting Diodes.

Chiral perovskites play a pivotal role in spintronics and optoelectronic systems attributed to their chiral-induced spin selectivity (CISS) effect. Specifically, they allow for spin-polarized charge transport in spin light-emitting diodes (LEDs), yielding circularly polarized electroluminescence at room temperature without external magnetic fields. However, chiral lead bromide-based perovskites have yet to achieve high-performance green emissive spin-LEDs, owing to limited CISS effects and charge transport. Herein, we employ dimensional regulation and Sn2+-doping to optimize chiral bromide-based perovskite architecture for green emissive spin-LEDs. The optimized (PEA)x(S/R-PRDA)2-xSn0.1Pb0.9Br4 chiral perovskite film exhibits an enhanced CISS effect, higher hole mobility, and better energy level alignment with the emissive layer. These improvements allow us to fabricate green emissive spin-LEDs with an external quantum efficiency (EQE) of 5.7% and an asymmetry factor |gCP-EL| of 1.1 × 10-3. This work highlights the importance of tailored perovskite architectures and doping strategies in advancing spintronics for optoelectronic applications.

DeepQs: Local quality assessment of cryo-EM density map by deep learning map-model fit score.

Cryogenic electron microscopy maps are valuable for determining macromolecule structures. A proper quality assessment method is essential for cryo-EM map selection or revision. This article presents DeepQs, a novel approach to estimate local quality for 3D cryo-EM density maps, using a deep-learning algorithm based on map-model fit score. DeepQs is a parameter-free method for users and incorporates structural information between map and its related atomic model into well-trained models by deep learning. More specifically, the DeepQs approach leverages the interplay between map and atomic model through predefined map-model fit score, Q-score. DeepQs can get close results to the ground truth map-model fit scores with only cryo-EM map as input. In experiments, DeepQs demonstrates the lowest root mean square error with standard method Fourier shell correlation metric and high correlation with map-model fit score, Q-score, when compared with other local quality estimation methods in high-resolution dataset (<=5 Å). DeepQs can also be applied to evaluate the quality of the post-processed maps. In both cases, DeepQs runs faster by using GPU acceleration. Our program is available at http://www.csbio.sjtu.edu.cn/bioinf/DeepQs for academic use.

HHL1 and SOQ1 synergistically regulate nonphotochemical quenching in Arabidopsis.

Nonphotochemical quenching (NPQ) is an important photoprotective mechanism that quickly dissipates excess light energy as heat. NPQ can be induced in a few seconds to several hours; most studies of this process have focused on the rapid induction of NPQ. Recently, a new, slowly induced form of NPQ, called qH, was found during the discovery of the quenching inhibitor suppressor of quenching 1 (SOQ1). However, the specific mechanism of qH remains unclear. Here, we found that hypersensitive to high light 1 (HHL1)-a damage repair factor of photosystem II-interacts with SOQ1. The enhanced NPQ phenotype of the hhl1 mutant is similar to that of the soq1 mutant, which is not related to energy-dependent quenching or other known NPQ components. Furthermore, the hhl1 soq1 double mutant showed higher NPQ than the single mutants, but its pigment content and composition were similar to those of the wildtype. Overexpressing HHL1 decreased NPQ in hhl1 to below wildtype levels, whereas NPQ in hhl1 plants overexpressing SOQ1 was lower than that in hhl1 but higher than that in the wildtype. Moreover, we found that HHL1 promotes the SOQ1-mediated inhibition of plastidial lipoprotein through its von Willebrand factor type A domain. We propose that HHL1 and SOQ1 synergistically regulate NPQ.

High light triggers flavonoid and polysaccharide synthesis through DoHY5-dependent signaling in Dendrobium officinale.

Dendrobium officinale is edible and has medicinal and ornamental functions. Polysaccharides and flavonoids, including anthocyanins, are important components of D. officinale that largely determine the nutritional quality and consumer appeal. There is a need to study the molecular mechanisms regulating anthocyanin and polysaccharide biosynthesis to enhance D. officinale quality and its market value. Here, we report that high light (HL) induced the accumulation of polysaccharides, particularly mannose, as well as anthocyanin accumulation, resulting in red stems. Metabolome and transcriptome analyses revealed that most of the flavonoids showed large changes in abundance, and flavonoid and polysaccharide biosynthesis was significantly activated under HL treatment. Interestingly, DoHY5 expression was also highly induced. Biochemical analyses demonstrated that DoHY5 directly binds to the promoters of DoF3H1 (involved in anthocyanin biosynthesis), DoGMPP2, and DoPMT28 (involved in polysaccharide biosynthesis) to activate their expression, thereby promoting anthocyanin and polysaccharide accumulation in D. officinale stems. DoHY5 silencing decreased flavonoid- and polysaccharide-related gene expression and reduced anthocyanin and polysaccharide accumulation, whereas DoHY5 overexpression had the opposite effects. Notably, naturally occurring red-stemmed D. officinale plants similarly have high levels of anthocyanin and polysaccharide accumulation and biosynthesis gene expression. Our results reveal a previously undiscovered role of DoHY5 in co-regulating anthocyanin and polysaccharide biosynthesis under HL conditions, improving our understanding of the mechanisms regulating stem color and determining nutritional quality in D. officinale. Collectively, our results propose a robust and simple strategy for significantly increasing anthocyanin and polysaccharide levels and subsequently improving the nutritional quality of D. officinale.

Spontaneous Degradation of the "Forever Chemicals" Perfluoroalkyl and Polyfluoroalkyl Substances (PFASs) on Water Droplet Surfaces.

Because of their innate chemical stability, the ubiquitous perfluoroalkyl and polyfluoroalkyl substances (PFASs) have been dubbed "forever chemicals" and have attracted considerable attention. However, their stability under environmental conditions has not been widely verified. Herein, perfluorooctanoic acid (PFOA), a widely used and detected PFAS, was found to be spontaneously degraded in aqueous microdroplets under room temperature and atmospheric pressure conditions. This unexpected fast degradation occurred via a unique multicycle redox reaction of PFOA with interfacial reactive species on the droplet surface. Similar degradation was observed for other PFASs. This study extends the current understanding of the environmental fate and chemistry of PFASs and provides insight into aid in the development of effective methods for removing PFASs.

The Enhancement Mechanism of Different Single-Transition Metal Atomic Catalysts/Sulfur Cathode on High-Performance of Li-S Batteries.

Materials with various single-transition metal atoms dispersed in nitrogenated carbons (M─N─C, M = Fe, Co, and Ni) are synthesized as cathodes to investigate the electrocatalytic behaviors focusing on their enhancement mechanism for performance of Li-S batteries. Results indicate that the order of both electrocatalytic activity and rate capacity for the M─N─C catalysts is Co > Ni > Fe, and the Co─N─C delivers the highest capacity of 1100 mAh g-1 at 1 C and longtime stability at a decay rate of 0.05% per cycle for 1000 cycles, demonstrating excellent battery performance. Theoretical calculations for the first time reveal that M─N─N─C catalysts enable direct conversion of Li2 S6 to Li2 S rather than Li2 S4 to Li2 S by stronger adsorption with Li2 S6 , which also has an order of Co > Ni > Fe. And Co─N─C has the strongest adsorption energy, not only rendering the highest electrocatalytic activity, but also depressing the polysulfides' dissolution into electrolyte for the longest cycle life. This work offers an avenue to design the next generation of highly efficient sulfur cathodes for high-performance Li-S batteries, while shedding light on the fundamental insight of single metal atomic catalytic effects on Li-S batteries.

Acidified Nitrogen Self-Doped Porous Carbon with Superprotonic Conduction for Applications in Solid-State Proton Battery.

Solid proton electrolytes play a crucial role in various electrochemical energy storage and conversion devices. However, the development of fast proton conducting solid proton electrolytes at ambient conditions remains a significant challenge. In this study, a novel acidified nitrogen self-doped porous carbon material is presented that demonstrates exceptional superprotonic conduction for applications in solid-state proton battery. The material, designated as MSA@ZIF-8-C, is synthesized through the acidification of nitrogen-doped porous carbon, specifically by integrating methanesulfonic acid (MSA) into zeolitic imidazolate framework-derived nitrogen self-doped porous carbons (ZIF-8-C). This study reveals that MSA@ZIF-8-C achieves a record-high proton conductivity beyond 10-2  S cm-1 at ambient condition, along with good long-term stability, positioning it as a cutting-edge alternative solid proton electrolyte to the default aqueous H2 SO4 electrolyte in proton batteries.

Response of the root anatomical structure of Carex moorcroftii to habitat drought in the Western Sichuan Plateau of China.

MAIN CONCLUSION: The anatomical structures of Carex moorcroftii roots showing stronger plasticity during drought had a lower coefficient of variation in cell size in the same habitats, while those showing weaker plasticity had a higher coefficient of variation. The complementary relationship between these factors comprises the adaptation mechanism of the C. moorcroftii root to drought. To explore the effects of habitat drought on root anatomy of hygrophytic plants, this study focused on roots of C. moorcroftii. Five sample plots were set up along a soil moisture gradient in the Western Sichuan Plateau to collect experimental materials. Paraffin sectioning was used to obtain root anatomy, and one-way ANOVA, correlation analysis, linear regression analysis, and RDA ranking were applied to analyze the relationship between root anatomy and soil water content. The results showed that the root transverse section area, thickness of epidermal cells, exodermis and Casparian strips, and area of aerenchyma were significantly and positively correlated with soil moisture content (P < 0.01). The diameter of the vascular cylinder and the number and total area of vessels were significantly and negatively correlated with the soil moisture content (P < 0.01). The plasticity of the anatomical structures was strong for the diameter and area of the vascular cylinder and thickness of the Casparian strip and epidermis, while it was weak for vessel diameter and area. In addition, there was an asymmetrical relationship between the functional adaptation of root anatomical structure in different soil moisture and the variation degree of root anatomical structure in the same soil moisture. Therefore, the roots of C. moorcroftii can shorten the water transport distance from the epidermis to the vascular cylinder, increase the area of the vascular cylinder and the number of vessels, and establish a complementary relationship between the functional adaptation of root anatomical structure in different habitats and the variation degree of root anatomical structure in the same habitat to adapt to habitat drought. This study provides a scientific basis for understanding the response of plateau wetland plants to habitat changes and their ecological adaptation strategies. More scientific experimental methods should be adopted to further study the mutual coordination mechanisms of different anatomical structures during root adaptation to habitat drought for hygrophytic plants.

SIFLoc: a self-supervised pre-training method for enhancing the recognition of protein subcellular localization in immunofluorescence microscopic images.

With the rapid growth of high-resolution microscopy imaging data, revealing the subcellular map of human proteins has become a central task in the spatial proteome. The cell atlas of the Human Protein Atlas (HPA) provides precious resources for recognizing subcellular localization patterns at the cell level, and the large-scale annotated data enable learning via advanced deep neural networks. However, the existing predictors still suffer from the imbalanced class distribution and the lack of labeled data for minor classes. Thus, it is necessary to develop new methods for coping with these issues. We leverage the self-supervised learning protocol to address these problems. Especially, we propose a pre-training scheme to enhance the conventional supervised learning framework called SIFLoc. The pre-training is featured by a hybrid data augmentation method and a modified contrastive loss function, aiming to learn good feature representations from microscopic images. The experiments are performed on a large-scale immunofluorescence microscopic image dataset collected from the HPA database. Using the same deep neural networks as the classifier, the model pre-trained via SIFLoc not only outperforms the model without pre-training by a large margin but also shows advantages over the state-of-the-art self-supervised learning methods. Especially, SIFLoc improves the prediction accuracy for minor organelles significantly.

Learning protein subcellular localization multi-view patterns from heterogeneous data of imaging, sequence and networks.

Location proteomics seeks to provide automated high-resolution descriptions of protein location patterns within cells. Many efforts have been undertaken in location proteomics over the past decades, thereby producing plenty of automated predictors for protein subcellular localization. However, most of these predictors are trained solely from high-throughput microscopic images or protein amino acid sequences alone. Unifying heterogeneous protein data sources has yet to be exploited. In this paper, we present a pipeline called sequence, image, network-based protein subcellular locator (SIN-Locator) that constructs a multi-view description of proteins by integrating multiple data types including images of protein expression in cells or tissues, amino acid sequences and protein-protein interaction networks, to classify the patterns of protein subcellular locations. Proteins were encoded by both handcrafted features and deep learning features, and multiple combining methods were implemented. Our experimental results indicated that optimal integrations can considerately enhance the classification accuracy, and the utility of SIN-Locator has been demonstrated through applying to new released proteins in the human protein atlas. Furthermore, we also investigate the contribution of different data sources and influence of partial absence of data. This work is anticipated to provide clues for reconciliation and combination of multi-source data for protein location analysis.

Accurate flexible refinement for atomic-level protein structure using cryo-EM density maps and deep learning.

With the rapid progress of deep learning in cryo-electron microscopy and protein structure prediction, improving the accuracy of the protein structure model by using a density map and predicted contact/distance map through deep learning has become an urgent need for robust methods. Thus, designing an effective protein structure optimization strategy based on the density map and predicted contact/distance map is critical to improving the accuracy of structure refinement. In this article, a protein structure optimization method based on the density map and predicted contact/distance map by deep-learning technology was proposed in accordance with the result of matching between the density map and the initial model. Physics- and knowledge-based energy functions, integrated with Cryo-EM density map data and deep-learning data, were used to optimize the protein structure in the simulation. The dynamic confidence score was introduced to the iterative process for choosing whether it is a density map or a contact/distance map to dominate the movement in the simulation to improve the accuracy of refinement. The protocol was tested on a large set of 224 non-homologous membrane proteins and generated 214 structural models with correct folds, where 4.5% of structural models were generated from structural models with incorrect folds. Compared with other state-of-the-art methods, the major advantage of the proposed methods lies in the skills for using density map and contact/distance map in the simulation, as well as the new energy function in the re-assembly simulations. Overall, the results demonstrated that this strategy is a valuable approach and ready to use for atomic-level structure refinement using cryo-EM density map and predicted contact/distance map.

circRNA-binding protein site prediction based on multi-view deep learning, subspace learning and multi-view classifier.

Circular RNAs (circRNAs) generally bind to RNA-binding proteins (RBPs) to play an important role in the regulation of autoimmune diseases. Thus, it is crucial to study the binding sites of RBPs on circRNAs. Although many methods, including traditional machine learning and deep learning, have been developed to predict the interactions between RNAs and RBPs, and most of them are focused on linear RNAs. At present, few studies have been done on the binding relationships between circRNAs and RBPs. Thus, in-depth research is urgently needed. In the existing circRNA-RBP binding site prediction methods, circRNA sequences are the main research subjects, but the relevant characteristics of circRNAs have not been fully exploited, such as the structure and composition information of circRNA sequences. Some methods have extracted different views to construct recognition models, but how to efficiently use the multi-view data to construct recognition models is still not well studied. Considering the above problems, this paper proposes a multi-view classification method called DMSK based on multi-view deep learning, subspace learning and multi-view classifier for the identification of circRNA-RBP interaction sites. In the DMSK method, first, we converted circRNA sequences into pseudo-amino acid sequences and pseudo-dipeptide components for extracting high-dimensional sequence features and component features of circRNAs, respectively. Then, the structure prediction method RNAfold was used to predict the secondary structure of the RNA sequences, and the sequence embedding model was used to extract the context-dependent features. Next, we fed the above four views' raw features to a hybrid network, which is composed of a convolutional neural network and a long short-term memory network, to obtain the deep features of circRNAs. Furthermore, we used view-weighted generalized canonical correlation analysis to extract four views' common features by subspace learning. Finally, the learned subspace common features and multi-view deep features were fed to train the downstream multi-view TSK fuzzy system to construct a fuzzy rule and fuzzy inference-based multi-view classifier. The trained classifier was used to predict the specific positions of the RBP binding sites on the circRNAs. The experiments show that the prediction performance of the proposed method DMSK has been improved compared with the existing methods. The code and dataset of this study are available at https://github.com/Rebecca3150/DMSK.

MDGF-MCEC: a multi-view dual attention embedding model with cooperative ensemble learning for CircRNA-disease association prediction.

Circular RNA (circRNA) is closely involved in physiological and pathological processes of many diseases. Discovering the associations between circRNAs and diseases is of great significance. Due to the high-cost to verify the circRNA-disease associations by wet-lab experiments, computational approaches for predicting the associations become a promising research direction. In this paper, we propose a method, MDGF-MCEC, based on multi-view dual attention graph convolution network (GCN) with cooperative ensemble learning to predict circRNA-disease associations. First, MDGF-MCEC constructs two disease relation graphs and two circRNA relation graphs based on different similarities. Then, the relation graphs are fed into a multi-view GCN for representation learning. In order to learn high discriminative features, a dual-attention mechanism is introduced to adjust the contribution weights, at both channel level and spatial level, of different features. Based on the learned embedding features of diseases and circRNAs, nine different feature combinations between diseases and circRNAs are treated as new multi-view data. Finally, we construct a multi-view cooperative ensemble classifier to predict the associations between circRNAs and diseases. Experiments conducted on the CircR2Disease database demonstrate that the proposed MDGF-MCEC model achieves a high area under curve of 0.9744 and outperforms the state-of-the-art methods. Promising results are also obtained from experiments on the circ2Disease and circRNADisease databases. Furthermore, the predicted associated circRNAs for hepatocellular carcinoma and gastric cancer are supported by the literature. The code and dataset of this study are available at https://github.com/ABard0/MDGF-MCEC.

De novo drug design by iterative multiobjective deep reinforcement learning with graph-based molecular quality assessment.

MOTIVATION: Generating molecules of high quality and drug-likeness in the vast chemical space is a big challenge in the drug discovery. Most existing molecule generative methods focus on diversity and novelty of molecules, but ignoring drug potentials of the generated molecules during the generation process. RESULTS: In this study, we present a novel de novo multiobjective quality assessment-based drug design approach (QADD), which integrates an iterative refinement framework with a novel graph-based molecular quality assessment model on drug potentials. QADD designs a multiobjective deep reinforcement learning pipeline to generate molecules with multiple desired properties iteratively, where a graph neural network-based model for accurate molecular quality assessment on drug potentials is introduced to guide molecule generation. Experimental results show that QADD can jointly optimize multiple molecular properties with a promising performance and the quality assessment module is capable of guiding the generated molecules with high drug potentials. Furthermore, applying QADD to generate novel molecules binding to a biological target protein DRD2 also demonstrates the algorithm's efficacy. AVAILABILITY AND IMPLEMENTATION: QADD is freely available online for academic use at https://github.com/yifang000/QADD or http://www.csbio.sjtu.edu.cn/bioinf/QADD.