Objectives: Coronavirus disease 2019 (COVID-19) is sweeping the globe and has resulted in infections in millions of people. Patients with COVID-19 face a high fatality risk once symptoms worsen; therefore, early identification of severely ill patients can enable early intervention, prevent disease progression, and help reduce mortality. This study aims to develop an artificial intelligence-assisted tool using computed tomography (CT) imaging to predict disease severity and further estimate the risk of developing severe disease in patients suffering from COVID-19. Materials and Methods: Initial CT images of 408 confirmed COVID-19 patients were retrospectively collected between January 1, 2020 and March 18, 2020 from hospitals in Honghu and Nanchang. The data of 303 patients in the People's Hospital of Honghu were assigned as the training data, and those of 105 patients in The First Affiliated Hospital of Nanchang University were assigned as the test dataset. A deep learning based-model using multiple instance learning and residual convolutional neural network (ResNet34) was developed and validated. The discrimination ability and prediction accuracy of the model were evaluated using the receiver operating characteristic curve and confusion matrix, respectively. Results: The deep learning-based model had an area under the curve (AUC) of 0.987 (95% confidence interval [CI]: 0.968-1.00) and an accuracy of 97.4% in the training set, whereas it had an AUC of 0.892 (0.828-0.955) and an accuracy of 81.9% in the test set. In the subgroup analysis of patients who had non-severe COVID-19 on admission, the

Summary Yield in rice is determined mainly by panicle architecture. Using map‐based cloning, we identified an R2R3 MYB transcription factor REGULATOR OF GRAIN NUMBER1 (RGN1) affecting grain number and panicle architecture. Mutation of RGN1 caused an absence of lateral grains on secondary branches. We demonstrated that RGN1 controls lateral grain formation by regulation of LONELY GUY (LOG) expression, thus controlling grain number and shaping panicle architecture. A novel favourable allele, RGN1C, derived from the Or‐I group in wild rice affected panicle architecture by means longer panicles. Identification of RGN1 provides a theoretical basis for understanding the molecular mechanism of lateral grain formation in rice; RGN1 will be an important gene resource for molecular breeding for higher yield.

Fe‐N‐doped graphitic carbon materials exhibit high efficiency and durability for oxygen reduction reaction (ORR). Although iron has relatively low price, the precursors for carbon and nitrogen used in previous studies have relatively high cost. Here reported is the preparation of highly efficient Fe‐N/C‐based ORR electrocatalysts by use of low‐cost urea as the precursors. Fe‐N/C‐based hybrids are prepared through a two‐step pyrolysis. During the first‐step pyrolysis, the precursors convert into g‐C3N4 with Fe located into the sixfold cavities, which ensures the relatively uniform distribution of Fe. The second‐step pyrolysis converts Fe‐g‐C3N4 into Fe‐N/C‐based hybrids which contain multiple types of active components, Fe moieties (FeCxNy or FeNx), Fe and Fe3N nanoparticles, for ORR. The obtained Fe‐N/C‐based hybrids display a superior electrocatalytic performance for ORR with an onset potential of 0.940 V and half‐wave potential of 0.810 V versus reversible hydrogen electrode, which are comparable to those of Pt/C at the same catalyst loading. The hybrids show higher tolerance to methanol and much greater long‐term stability than commercial Pt/C. The findings provide a cost‐effective approach for the preparation of high efficient and stable electrocatalysts for ORR and will be very helpful to the development of electrochemical energy storage and conversion.

Summary Rice ( Oryza sativa L.) cultivars harbour morphological and physiological traits different from those of wild rice ( O. rufipogon Griff.), but the molecular mechanisms underlying domestication remain controversial. Here, we show that awn and long grain traits in the near‐isogenic NIL ‐ GLA are separately controlled by variations within the GLA ( Grain Length and Awn Development ) gene, a new allele of GAD 1 / RAE 2 , which encodes one member of the EFPL (epidermal patterning factor‐like protein) family. Haplotype analyses and transgenic studies revealed that InDel1 (variation for grain length, VGL ) in the promoter region of GLA ( GLA VGL ) increases grain length by promoting transcription of GLA . Absence of InDel3 (variation for awn formation, VA ) in the coding region ( CDS ) of GLA ( GLA va ) results in short awn or no awn phenotypes. Analyses of minimum spanning trees and introgression regions demonstrated that An‐1 , an important gene for awn formation, was preferentially domesticated and its mutation to an‐1 was followed by GLA and An‐2 . Gene flow then occurred between the evolved japonica and indica populations. Quality analysis showed that GLA causes poor grain quality. During genetic improvement, awnlessness was selected in ssp. indica , whereas short–grained and awnless phenotypes with good quality were selected in japonica . Our findings facilitate an understanding of rice domestication and provide a favourable allele for rice breeding.