Corynebacterium zhongnanshanii sp. nov. isolated from trachea of Marmota himalayana, Corynebacterium lujinxingii sp. nov. and Corynebacterium wankanglinii sp. nov. from human faeces.

Six novel facultatively anaerobic, Gram-stain-positive, rod-shaped, non-haemolytic bacteria (zg-320T/zg-336, zg-917T/zg-910 and zg-913T/zg-915) isolated from animal tissues and human faeces were found to belong to the genus Corynebacterium based on the phylogenetic analyses of 16S rRNA gene and 262 core genes set. Based on the greatest degree of 16S rRNA similarity, zg-320T/zg-336 had the highest 16S rRNA gene similarity to Corynebacterium falsenii DSM 44353T (97.51 %), zg-917T/zg-910 to Corynebacterium coyleae DSM 44184T (98.68 %), and zg-913T/zg-915 to Corynebacterium afermentans subsp. lipophilum CIP 103500T (98.79 %). The three novel type strains had a relatively high DNA G+C content (61.2-64.4 mol%), low DNA relatedness and ANI values with their respective neighbours: 23.5/72.7 %, 25.0/72.3%and 22.6/73.1 % (zg-320T vs. Corynebacterium auriscanis CIP 106629T, Corynebacterium resistens DSM 45100T and Corynebacterium suicordis DSM 45110T); 24.4/82.3% and 23.7/81.3 % (zg-917T vs. C. coyleae DSM 44184T and Corynebacterium jeddahense JCBT); 26.8/83.7% and 27.7/84.4 % (zg-913T vs. Corynebacterium mucifaciens ATCC 700355T and C. afermentans subsp. lipophilum CCUG 32105T). The three novel species had C16 : 0, C18 : 0, C18 : 1 ω9c and C18 : 0 ante/C18 : 2 ω6,9c as the major cellular fatty acids; MK-8(H2) in strain zg-917T and MK-9(H2) in strains zg-320T and zg-913T were found to be the major respiratory quinones. For the three novel species, the detected major polar lipids included diphosphatidylglycerol, phosphatidyl inositol mannoside, phosphatidylglycerol and phosphatidylinositol, the cell-wall peptidoglycan was based on meso-DAP, and the whole-cell sugars mainly included ribose, arabinose and galactose. The three novel species grew optimally at 35-37 °C, 0.5 % (w/v) NaCl and pH 7.0-8.0; notably, they were tolerant of 10.5 % (w/v) NaCl. Based on the results of these comprehensive analyses, three novel species in the genus Corynebacterium are proposed, aptly named Corynebacterium zhongnanshanii sp. nov. (zg-320T = GDMCC 1.1719T = JCM 34106T), Corynebacterium lujinxingii sp. nov. (zg-917T = GDMCC 1.1707T = JCM 34094T) and Corynebacterium wankanglinii sp. nov. (zg-913T = GDMCC 1.1706T = JCM 34398T).

Identification of Haloactinobacterium kanbiaonis sp. nov. and Ruania zhangjianzhongii sp. nov., two novel species of the family Ruaniaceae isolated from faeces of bats (Hipposideros spp.).

Two pairs of aerobic, Gram-stain-positive, rod-shaped strains (HY164T/HY044, HY168T/HY211) were isolated from bat faecal samples. Strains HY164T and HY044 were motile with a polar flagellum, and had 16S rRNA gene similarity of 95.1-98.6 % to Haloactinobacterium album YIM 93306T and Haloactinobacterium glacieicola T3246-1T; strains HY168T and HY211 were most similar to Ruania albidiflava DSM 18029T (96.6 %). Phylogenetic trees based on 16S rRNA gene and whole genome sequences revealed affiliation of strains HY164T and HY168T to the family Ruaniaceae, representing novel lineages in the genera Haloactinobacterium and Ruania, respectively, which was also supported by the results for average nucleotide identity (ANI) and digital DNA-DNA hybridization (dDDH). For all isolates, the principal cellular fatty acids were anteiso-C15 : 0 and iso-C14 : 0. HY164T and HY168T had MK-8(H4) as the predominant isoprenoid quinone, diphosphatidylglycerol, phosphatidylglycerol, several unidentified phospholipids and glycolipids as common polar lipids while the latter strain additionally contained one unidentified aminophospholipid and one unidentified phosphoglycolipid. Besides sharing alanine, glutamic acid and lysine with HY164T, HY168T additionally contained 2,4-diaminobutyric acid in the cell-wall peptidoglycan. The whole-cell sugars of HY164T were ribose and rhamnose, while HY168T only included the latter. The DNA G+C contents of HY164T and HY168T were 71.0 and 69.1 mol%, respectively. Combining the polyphasic taxonomic data, HY164T (=CGMCC 4.7606T=JCM 33464T) is classified as representing a novel species of the genus Haloactinobacterium with the proposed name Haloactinobacterium kanbiaonis sp. nov., and HY168T (=CGMCC 1.16970T=JCM 33465T) is proposed to represent a novel species of the genus Ruania with the name Ruania zhangjianzhongii sp. nov.

Nocardioides dongkuii sp. nov. and Nocardioides lijunqiniae sp. nov., isolated from faeces of Tibetan antelope (Pantholops hodgsonii) and leaves of dandelion (Taraxacum officinale), respectively, on the Qinghai-Tibet Plateau.

In the present study, four bacterial strains, two (S-713T and 406) isolated from faecal samples of Tibetan antelopes and the other two (S-531T and 1598) from leaves of dandelion collected on the Qinghai-Tibet Plateau of PR China, were analysed using a polyphasic approach. All four isolates were aerobic, rod-shaped, non-motile, oxidase-negative, Gram-stain-positive and catalase-positive. According to four phylogenetic trees, strain pairs S-713T/406 and S-531T/1598 form two independent branches belonging to the genus Nocardioides, and are closest to Nocardioides lianchengensis, Nocardioides dokdonensis, Nocardioides salarius, Nocardioides marinisabuli, Nocardioides psychrotolerans and Nocardioides szechwanensis. Although sharing MK8-(H4) as their major isoprenoid quinone, strains S-713T and S-531T contained C18 : 1 ω9c (24.64 and 16.34 %) and iso-C16 : 0 (9.74 and 29.38 %), respectively, as their main fatty acids, with remarkable differences in their biochemical profiles but only slight ones in their optimal growth conditions. The chromosomes of strains S-713T and S-531T were 4 207 844 bp (G+C content, 73.0 mol%) and 4 809 817 bp (G+C content, 72.5 mol%), respectively. Collectively, the two strain pairs represent two separate novel species of the genus Nocardioides, for which the names Nocardioides dongkuii sp. nov. and Nocardioides lijunqiniae sp. nov. are proposed, with S-713T (=JCM 33698T=CGMCC 4.7660T) and S-531T (=JCM 33468T=CGMCC 4.7659T) as the respective type strains.

Source-Guided Target Feature Reconstruction for Cross-Domain Classification and Detection.

Existing cross-domain classification and detection methods usually apply a consistency constraint between the target sample and its self-augmentation for unsupervised learning without considering the essential source knowledge. In this paper, we propose a Source-guided Target Feature Reconstruction (STFR) module for cross-domain visual tasks, which applies source visual words to reconstruct the target features. Since the reconstructed target features contain the source knowledge, they can be treated as a bridge to connect the source and target domains. Therefore, using them for consistency learning can enhance the target representation and reduce the domain bias. Technically, source visual words are selected and updated according to the source feature distribution, and applied to reconstruct the given target feature via a weighted combination strategy. After that, consistency constraints are built between the reconstructed and original target features for domain alignment. Furthermore, STFR is connected with the optimal transportation algorithm theoretically, which explains the rationality of the proposed module. Extensive experiments on nine benchmarks and two cross-domain visual tasks prove the effectiveness of the proposed STFR module, e.g., 1) cross-domain image classification: obtaining average accuracy of 91.0%, 73.9%, and 87.4% on Office-31, Office-Home, and VisDA-2017, respectively; 2) cross-domain object detection: obtaining mAP of 44.50% on Cityscapes → Foggy Cityscapes, AP on car of 78.10% on Cityscapes → KITTI, MR -2 of 8.63%, 12.27%, 22.10%, and 40.58% on COCOPersons → Caltech, CityPersons → Caltech, COCOPersons → CityPersons, and Caltech → CityPersons, respectively.

[Prognostic Value of IGF2BP3 Gene Expression Levels in Patients with Acute Myeloid Leukemia].

OBJECTIVE: To investigate the relationship between IGF2BP3 gene expression and prognosis in patients with acute myeloid leukemia (AML). METHODS: High throughput transcriptome sequencing was performed on bone marrow primary leukemia cells from 27 patients with AML in our center, the relationship between IGF2BP3 expression levels and clinical characteristics were analyzed and verify the samples from patients with newly treated AML and refractory AML. The expression level of IGF2BP3 gene were analyzed in 20 healthy subjects and 26 patients with AML. The expression of IGF2BP3 in two anthracycline-resistant cell lines (HL60/ADR, K562/ADR) was detected by RT-qPCR and Western blot, and the expression difference of IGF2BP3 was compared with that in sensitive cells (HL60, K562). The relationship between the expression level of IGF2BP3 in patients with AML and prognostic were analyzed through data analysis of 746 patients with AML, and the prognostic value of IGF2BP3 in AML was analyzed by multivariate Cox regression analysis. RESULTS: In the bone marrow primary leukemia cells of 27 AML patients in our center, the expression level of IGF2BP3 in patients with refractory AML was significantly higher than that in chemotherapy sensitive patients (P =0.0343). The expression of IGF2BP3 in leukemia patients with extramedullary infiltration (EMI) was significantly higher than that in AML patients without extramedullary infiltration (P =0.0049). Compared with healthy subjects (n=20), IGF2BP3 expression in AML patients (n=26) was higher (P =0.0009). The expression of IGF2BP3 mRNA in the anthracycline resistant cell lines (HL60/ADR, K562/ADR) was significantly higher than that in the sensitive cell lines (K562/ADR vs K562,P =0.0430; HL60/ADR vs HL60, P =0.7369). Western blot results showed that the expression of IGF2BP3 protein in mycin resistant cells was significantly higher than that in sensitive cells (P < 0.001). qPCR results showed that the expression level of IGF2BP3 mRNA in refractory AML patients was significantly higher than that in patients with chemotherapy sensitive (P =0.002). High expression of IGF2BP3 was associated with poor prognosis in AML (P < 0.05) in 3 large sample cohorts of AML patients. Univariate and multivariate prognostic analyses demonstrated that high expression of IGF2BP3 was significantly associated with shorter event-free survival (EFS, HR=1.887, P =0.024) and overall survival (OS, HR=1.619, P =0.016). CONCLUSION: The high expression of IGF2BP3 gene may be an important factor in the poor prognosis of AML, suggesting that IGF2BP3 gene may be a new molecular marker for the clinical prognosis evaluation and treatment strategy of AML.

[Venetoclax Combined with CACAG Regimen in the Treatment of Patients with Refractory/Relapse Acute Myeloid Leukemia: A Prospective Clinical Study].

OBJECTIVE: To investigate the efficacy and safety of Venetoclax combined with CACAG regimen in treatment of patients with refractory/relapse acute myeloid leukemia(R/R AML). METHODS: The study was a singlecenter prospective clinical trial. The enrolled patients met the criteria for R/R AML. Treatment included Azacidine(75 mg/m2，d 1-7), Ara-C (75-100 mg/m2, q12h, d 1-5), Aclacinomycin(20 mg d1,d3,d5), Chidamide(30 mg d1,d4), Venetoclax(100 mg d1, 200 mg d2, 400 mg d3-d14, in combination with Triazole Drug, reduced to 100 mg/d), and granulocyte colony-stimulating factor (300 µg /d until neutrophil recovery). The primary endpoint of observation was overall response rate after 1 course of treatment. RESULTS: A total of 19 patients were enrolled from January 2022 to April 2023. After 1 course of treatmen, the overall response rate was 81.3%(13/16), the CR rate was 68.8%(11/16), and the PR was 12.5%(2/16). Among the 11 patients who got CR/CRi, 8 cases achieved CRm (minimal residual disease negative CR) and 3 cases did not. As of March 27, 2023, the median follow-up time was 111(19-406) days. The six-month overall survival and progression-free survival rates were both 55.7%, the 1-year overall survival and progression-free survival rates were 46.4% and 47.7%, respectively. In addition, compared with the non-CRm group, CRm patients had a better PFS (377 days vs 111 days, P =0.046). Treatment-related adverse events were mainly 3-4 degrees of bone marrow suppression, complicated by various degrees of infection(n=12), hypokalemia(n=12) and hypocalcemia (n=10) and elevated liver enzymes (n=8), of which 3/4 degrees accounted for 47.4%(9/19). CONCLUSION: The Venetoclax combined with CACAG regimen is an effective salvage therapy for patients with R/R AML, with high remission rate and safety profile.

Dual Instance-Consistent Network for Cross-Domain Object Detection.

Cross-domain object detection aims to transfer knowledge from a labeled dataset to an unlabeled dataset. Most existing methods apply a unified embedding model to generate the tightly coupled source and target descriptions for domain alignment, leading to the destroyed feature distribution of the target domain because the embedding model is mainly controlled by the source domain. To reduce the representation bias of the target domain, we apply two independent networks to extract two types of discriminative descriptions with mutual consistency, i.e., a novel Dual Instance-Consistent Network (DICN) is proposed for cross-domain object detection. Especially, Dual Instance-Consistent Module containing the instance mutual consistency between Primary Network and Auxiliary Network is applied to align two domains, where Primary and Auxiliary Networks are used to obtain the source-specific and target-specific information, respectively. The instance mutual consistency consists of two terms: feature consistency and detection consistency, which is applied to align the instance feature and the output of detection head, respectively. With the instance mutual consistency, optimizing the Primary (Auxiliary) Network only with source (target) images by fixing the Auxiliary (Primary) Network can generate the source(target)-specific description. Extensive experiments on several benchmarks demonstrate the effectiveness of the proposed DICN, e.g., obtaining mAP of 44.10% for Cityscapes → Foggy Cityscapes, AP on car of 76.50% for Cityscapes → KITTI, MR -2 of 8.87%, 12.66%, 22.27%, and 42.06% for COCOPersons → Caltech, CityPersons → Caltech, COCOPersons → CityPersons, and Caltech → CityPersons, respectively.

Description of Ornithinimicrobium cryptoxanthini sp. nov., a Novel Actinomycete Producing ß-cryptoxanthin Isolated from the Tongtian River Sediments.

Two novel Gram-stain-positive, aerobic, non-motile, and yellow-pigmented, irregular rod-shaped bacteria (JY.X269 and JY.X270T) were isolated from the near-surface sediments of river in Qinghai Province, P. R. China (32°37'13″N, 96°05'37″E) in July 2019. Both strains were shown to grow at 15-35 °C and pH 7.0-10.0, and in the presence of 0-6.0% (w/v) NaCl. The 16S rRNA gene sequence analysis showed that the isolates were closely related to Ornithinimicrobium cavernae CFH 30183 T (98.6-98.8% 16S rRNA gene sequence similarity), O. ciconiae H23M54T (98.5-98.6%) and O. murale 01-Gi-040T (98.3-98.5%). The phylogenetic and phylogenomic trees based on the 16S rRNA gene and 537 core gene sequences, respectively, revealed that the two strains formed a distinct cluster with the above three species. The digital DNA-DNA hybridization (dDDH) and average nucleotide identity (ANI) values between our two isolates (JY.X269 and JY.X270T) and other Ornithinimicrobium species were within the ranges of 19.0-23.9% and 70.8-80.4%, respectively, all below the respective recommended 70.0% and 95-96% cutoff point. Furthermore, the major cellular fatty acids (> 10.0%) of strains JY.X269 and JY.X270T were iso-C15:0, iso-C16:0, and summed feature 9. Strain JY.X270T contained MK-8(H4) and ornithine as the predominant menaquinone and diagnostic diamino acid component within the cell wall teichoic acids. ß-cryptoxanthin (C40H56O) can be extracted from strain JY.X270T, and its content is 6.3 µg/ml. Based on results from the phylogenetic, chemotaxonomic, and phenotypic analyses, the two strains could be classified as a novel species of the genus Ornithinimicrobium, for which the name Ornithinimicrobium cryptoxanthini sp. nov. is proposed (type strain JY.X270T = CGMCC 1.19147T = JCM 34882T).

A novel scoring system for the quantitative prediction of prognosis in acute myeloid leukemia.

Background: Acute myeloid leukemia (AML) is a heterogeneous hematopoietic malignancy. Patient prognosis cannot be accurately assessed in National Comprehensive Cancer Network (NCCN) risk stratification subgroups based on the current criteria. This study aimed to develop a novel prognostic score model for the quantitative prediction of prognosis in AML. Results: We developed a prognostic risk scoring model of AML using differentially expressed genes to predict prognosis in patients with AML. Furthermore, we evaluated the effectiveness and clinical significance of this prognostic model in 4 AML cohorts and 905 patients with AML. A prognostic risk scoring model of AML containing eight prognosis-related genes was constructed using a multivariate Cox regression model. The model had a higher predictive value for the prognosis of AML in the training and validation sets. In addition, patients with lower scores had significantly better overall survival (OS) and even-free survival (EFS) than those with higher scores among patients with intermediate-risk AML according to the NCCN guidelines, indicating that the model could be used to further predict the prognosis of the intermediate-risk AML populations. Similarly, patients with high scores had remarkably poor OS and EFS in the normal-karyotype populations, indicating that the scoring model had an excellent predictive performance for patients with AML having normal karyotype. Conclusions: Our study provided an individualized prognostic risk score model that could predict the prognosis of patients with AML.

Emergence of an ancient and pathogenic mammarenavirus.

Emerging zoonoses of wildlife origin caused by previously unknown agents are one of the most important challenges for human health. The Qinghai-Tibet Plateau represents a unique ecological niche with diverse wildlife that harbours several human pathogens and numerous previously uncharacterized pathogens. In this study, we identified and characterized a novel arenavirus (namely, plateau pika virus, PPV) from plateau pikas (Ochotona curzoniae) on the Qinghai-Tibet Plateau by virome analysis. Isolated PPV strains could replicate in several mammalian cells. We further investigated PPV pathogenesis using animal models. PPV administered via an intraventricular route caused trembling and sudden death in IFNαßR-/- mice, and pathological inflammatory lesions in brain tissue were observed. According to a retrospective serological survey in the geographical region where PPV was isolated, PPV-specific IgG antibodies were detected in 8 (2.4%) of 335 outpatients with available sera. Phylogenetic analyses revealed that this virus was clearly separated from previously reported New and Old World mammarenaviruses. Under the co-speciation framework, the estimated divergence time of PPV was 77-88 million years ago (MYA), earlier than that of OW and NW mammarenaviruses (26-34 MYA).

Transcriptome-Wide Analysis of RNA N6-Methyladenosine Modification in Adriamycin-Resistant Acute Myeloid Leukemia Cells.

Acute myeloid leukemia (AML) is one of the most aggressive hematopoietic malignancies. Patients still suffer from refractory/relapsed disease after anthracycline-based therapy, which leads to a poor prognosis. N6-Methyladenosine (m6A) is the most abundant post-transcriptional modification in eukaryotes, the imbalance of which is reported to be associated with various pathological processes, including drug resistance. However, the relationship between m6A modification and drug resistance has not been well defined in AML. In this study, we analyzed the sequencing data of HL60 and its Adriamycin-resistant cell line HL60/ADR. We found a total of 40,550 m6A-methylated peaks, representing 15,640 genes in HL60, and 38,834 m6A-methylated peaks, representing 15,285 genes in HL60/ADR. KEGG pathway analysis showed that pathways were enriched in the FoxO signaling pathway, p53 signaling pathway, and Notch signaling pathway. MeRIP-seq results showed that the fold enrichment of the global m6A level in HL60/ADR was higher than that in HL60, and dot blot assay results indicated that the global m6A level was elevated in HL60/ADR cells compared with that in HL60 cells. Further analysis revealed that the expression level of METTL3 was elevated in HL60/ADR cells compared with that in HL60 cells. After a combined treatment of STM2457 (an inhibitor of METTL3) and Adriamycin, the proliferation of HL60/ADR was inhibited. Thus, we hypothesized that the abnormality of m6A modification played an important role in Adriamycin-resistant AML.

Flaviflexus equikiangi sp. nov. isolated from faeces of Equus kiang (Tibetan wild ass) and carrying a class 1 integron gene cassette in its genome.

Two Gram-stain-positive, catalase-negative, non-spore-forming, cocci-shaped strains (dk850T and JY899) were isolated from the feces of Equus kiang in the Qinghai-Tibet Plateau of China. 16S rRNA gene sequence-based phylogenetic analyses showed that strains dk850T and JY899 belong to the genus Flaviflexus, closest to F. salsibiostraticola KCTC 33148T, F. ciconiae KCTC 49253T and F. huanghaiensis H5T. The DNA G + C content of strain dk850T was 62.9%. The digital DNA-DNA hybridization values of strain dk850T with the closely related species were below the 70% threshold for species demarcation. The two strains grew best at 28°C on brain heart infusion (BHI) agar with 5% sheep blood. All strains had C18:1ω9c and C16:0 as the major cellular fatty acids. MK-9(H4) was the major menaquinone in strain dk850T. The major polar lipids included diphosphatidylglycerol and an unidentified phospholipid. Strains dk850T and JY899 were identified as carrying a class 1 integron containing the aminoglycoside resistance gene aadA11, both strains were resistant to spectinomycin and streptomycin. Based on several lines of evidence from phenotypic and phylogenetic analyses, strains dk850T and JY899 represent a novel species of the genus Flaviflexus, for which the name Flaviflexus equikiangi sp. nov. is proposed. The type strain is dk850T(= CGMCC 1.16593T = JCM 33598T).

Changpingibacter yushuensis gen. nov., sp. nov., isolated from fluvial sediment in Qinghai Tibet Plateau of China.

Two facultatively anaerobic, short rod-shaped, non-motile, Gram-stain-positive, unknown bacterial strains (JY-X040T and JY-X174) were isolated from fluvial sediments of Tongtian River in Yushu Tibetan Autonomous Prefecture, Qinghai province, China. Cells formed translucent, gray, round and convex colonies, with a diameter of less than 0.5 mm after 5 days of incubation at 30°C on brain heart infusion-5% sheep blood agar. The 16S rRNA gene sequence similarity between strain JY-X040T and Fudania jinshanensis 313T is 93.87%. In the four phylogenetic trees constructed based on the 16S rRNA gene and 423 core genes, the two isolates form an independent branch, phylogenetically closest to F. jinshanensis 313T, but could not be classified as a member of the genus Fudania or any other genus of the family Arcanobacteriaceae. The DNA G + C content of strain JY-X040T was 57.8%. Calculation results of average nucleotide identity, digital DNA-DNA hybridization value and amino acid identity between strain JY-X040T and F. jinshanensis 313T are 69.9%, 22.9%, and 64.1%. The major cellular fatty acids were C16:0 (23%) and C18:1ω9c (22%). The cell-wall peptidoglycan type was A5α (L-Lys-L-Ala-L-Lys-D-Glu). The polar lipids comprised diphosphatidylglycerol, phosphatidylglycerol, phosphatidylinositol, phosphatidylinositol mannoside and four unidentified components. The whole-cell sugars contained rhamnose and ribose. MK-10(H4) was the sole respiratory quinone. The minimum inhibitory concentration of streptomycin was 32 µg/ml. All physiological, biochemical, chemotaxonomic and genomic characteristics support that strains JY-X040T and JY-X174 represent members of a novel species in a new genus, Changpingibacter yushuensis gen. nov., sp. nov. The type strain is JY-X040T (GDMCC 1.1996T = KCTC 49514T).

Reduced Risk of Chronic Graft-Versus-Host Disease (cGVHD) by Rabbit Anti-Thymocyte Globulin (ATG) in Patients Undergoing Matched Sibling Donor Transplantation in Hematological Malignancies.

BACKGROUND With the addition of anti-thymocyte globulin (ATG) to GVHD prophylaxis in patients undergoing transplantation of peripheral blood stem cells (PBSCT), the incidence of cGVHD decreases. However, the optimal dose and timing of ATG remain undetermined. MATERIAL AND METHODS In this historical controlled trial, data from 85 patients who had hematological malignancies and underwent matched sibling donor (MSD)-PBSCT were used to analyze the effectiveness of rabbit ATG (rATG) for prophylaxis of GVHD. Forty patients received 5 mg/kg rATG used for days -5 to -2, and 45 patients did not receive ATG. RESULTS All patients had successful engraftment except for 2 in the non-ATG group, who had platelet engraftment failure. The 2-year cumulative incidence of chronic GVHD (cGVHD) in the ATG group versus non-ATG group was 19.3% (95% CI, 8.4-33.6%) versus 61.4% (95% CI, 45.4-73.9%) (P<0.001), and in those with moderate to severe cGVHD it was 11.0% (95% CI, 3.4-23.6%) versus 31.8% (95% CI, 18.8-45.6%) (P=0.029), respectively. The 2-year cumulative incidence of non-relapse mortality and relapse (CIR) were 0% versus 15.5% (95% CI, 6.8-27.5%) (P=0.018), and 53.3% (95% CI, 35.6-68.1%) versus 26.7% (95% CI, 14.9-40.0%) (P=0.019), respectively. No differences were found in other survival outcomes. In the multivariate analysis, ATG was an independent protective factor for moderate to severe cGVHD (HR=0.314, 95% CI, 0.103-0.958, P=0.042), and was an independent poor risk factor for CIR (HR=2.337, 95% CI, 1.133-4.822, P=0.022). CONCLUSIONS ATG in our strategy was effective for prophylaxis of cGVHD, whereas the relapse rate was increased in patients with rATG.

SAN: Selective Alignment Network for Cross-Domain Pedestrian Detection.

Cross-domain pedestrian detection, which has been attracting much attention, assumes that the training and test images are drawn from different data distributions. Existing methods focus on aligning the descriptions of whole candidate instances between source and target domains. Since there exists a giant visual difference among the candidate instances, aligning whole candidate instances between two domains cannot overcome the inter-instance difference. Compared with aligning the whole candidate instances, we consider that aligning each type of instances separately is a more reasonable manner. Therefore, we propose a novel Selective Alignment Network for cross-domain pedestrian detection, which consists of three components: a Base Detector, an Image-Level Adaptation Network, and an Instance-Level Adaptation Network. The Image-Level Adaptation Network and Instance-Level Adaptation Network can be regarded as the global-level and local-level alignments, respectively. Similar to the Faster R-CNN, the Base Detector, which is composed of a Feature module, an RPN module and a Detection module, is used to infer a robust pedestrian detector with the annotated source data. Once obtaining the image description extracted by the Feature module, the Image-Level Adaptation Network is proposed to align the image description with an adversarial domain classifier. Given the candidate proposals generated by the RPN module, the Instance-Level Adaptation Network firstly clusters the source candidate proposals into several groups according to their visual features, and thus generates the pseudo label for each candidate proposal. After generating the pseudo labels, we align the source and target domains by maximizing and minimizing the discrepancy between the prediction of two classifiers iteratively. Extensive evaluations on several benchmarks demonstrate the effectiveness of the proposed approach for cross-domain pedestrian detection.

Food-trade-associated COVID-19 outbreak from a contaminated wholesale food supermarket in Beijing.

The re-emerging outbreak of COVID-19 in Beijing, China, in the summer of 2020 originated from a SARS-CoV-2-infested wholesale food supermarket. We postulated that the Xinfadi market outbreak has links with food-trade activities. Our Susceptible to the disease, Infectious, and Recovered coupled Agent Based Modelling (SIR-ABM) analysis for studying the diffusion of SARS-CoV-2 particles suggested that the trade-distancing strategy effectively reduces the reproduction number (R0). The retail shop closure strategy reduced the number of visitors to the market by nearly half. In addition, the buy-local policy option reduced the infection by more than 70% in total. Therefore, retail closures and buy-local policies could serve as significantly effective strategies that have the potential to reduce the size of the outbreak and prevent probable outbreaks in the future.