BioByGANS: biomedical named entity recognition by fusing contextual and syntactic features through graph attention network in node classification framework.

BACKGROUND: Automatic and accurate recognition of various biomedical named entities from literature is an important task of biomedical text mining, which is the foundation of extracting biomedical knowledge from unstructured texts into structured formats. Using the sequence labeling framework and deep neural networks to implement biomedical named entity recognition (BioNER) is a common method at present. However, the above method often underutilizes syntactic features such as dependencies and topology of sentences. Therefore, it is an urgent problem to be solved to integrate semantic and syntactic features into the BioNER model. RESULTS: In this paper, we propose a novel biomedical named entity recognition model, named BioByGANS (BioBERT/SpaCy-Graph Attention Network-Softmax), which uses a graph to model the dependencies and topology of a sentence and formulate the BioNER task as a node classification problem. This formulation can introduce more topological features of language and no longer be only concerned about the distance between words in the sequence. First, we use periods to segment sentences and spaces and symbols to segment words. Second, contextual features are encoded by BioBERT, and syntactic features such as part of speeches, dependencies and topology are preprocessed by SpaCy respectively. A graph attention network is then used to generate a fusing representation considering both the contextual features and syntactic features. Last, a softmax function is used to calculate the probabilities and get the results. We conduct experiments on 8 benchmark datasets, and our proposed model outperforms existing BioNER state-of-the-art methods on the BC2GM, JNLPBA, BC4CHEMD, BC5CDR-chem, BC5CDR-disease, NCBI-disease, Species-800, and LINNAEUS datasets, and achieves F1-scores of 85.15%, 78.16%, 92.97%, 94.74%, 87.74%, 91.57%, 75.01%, 90.99%, respectively. CONCLUSION: The experimental results on 8 biomedical benchmark datasets demonstrate the effectiveness of our model, and indicate that formulating the BioNER task into a node classification problem and combining syntactic features into the graph attention networks can significantly improve model performance.

Genomic correlates of programmed cell death ligand 1 (PD-L1) expression in Chinese lung adenocarcinoma patients.

BACKGROUND: Although PD-L1 expression is a crucial predictive biomarker for immunotherapy, it can be influenced by many factors. METHODS: A total of 248 Chinese patients with lung adenocarcinoma was retrospectively identified. Data for clinical features, gene alternations, signaling pathways and immune signatures was analyzed among negative expression group (TPS < 1%, n = 124), intermediate expression group (1% ≤ TPS < 50%, n = 93), and high expression group (TPS ≥ 50%, n = 38). Clinical outcomes among different expression groups were also evaluated from public database. RESULTS: Firstly, high tumor mutation burden was significantly associated with high PD-L1 expression in these Chinese patients with lung adenocarcinoma. In addition, gene alternations including TP53, PRKDC, KMT2D, TET1 and SETD2 apparently occurred in high PD-L1 expression group. Moreover, pathway analysis showed that mutations involving in DDR pathway, TP53 pathway, cell-cycle pathway and NOTCH pathway were obviously varied among three PD-L1 expression groups. Besides, most of patients in high PD-L1 expression group from TCGA database were determined as high-grade immune subtypes (C2-C4), showing significant higher proportions of IFN-gamma, CD8+ T-cells, NK cells, NK CD56 dim cells, Th1 cells, Th2 cells (P < 0.0001). Moreover, SETD2 mutation slightly correlated with overall survival from MSKCC cohort (HR 1.92 [95%CI 0.90-4.10], P = 0.085), and the percentage of IFN-gamma was significantly higher in SETD2 mutant group than in wild-type group (P < 0.01). CONCLUSIONS: This study illustrated in-depth genomic correlates of PD-L1 expression in Chinese lung adenocarcinoma patients and relevant immune signatures from public database, which might interpret more potential molecular mechanisms for immunotherapy in NSCLC.

A novel species of ellipsoidal multicellular magnetotactic prokaryotes from Lake Yuehu in China.

Two morphotypes of multicellular magnetotactic prokaryotes (MMPs) have been identified: spherical (several species) and ellipsoidal (previously one species). Here, we report novel ellipsoidal MMPs that are ∼ 10 × 8 µm in size, and composed of about 86 cells arranged in six to eight interlaced circles. Each MMP was composed of cells that synthesized either bullet-shaped magnetite magnetosomes alone, or both bullet-shaped magnetite and rectangular greigite magnetosomes. They showed north-seeking magnetotaxis, ping-pong motility and negative phototaxis at a velocity up to 300 µm s(-1) . During reproduction, they divided along either their long- or short-body axes. For genetic analysis, we sorted the ellipsoidal MMPs with micromanipulation and amplified their genomes using multiple displacement amplification. We sequenced the 16S rRNA gene and found 6.9% sequence divergence from that of ellipsoidal MMPs, Candidatus Magnetananas tsingtaoensis and > 8.3% divergence from those of spherical MMPs. Therefore, the novel MMPs belong to different species and genus compared with the currently known ellipsoidal and spherical MMPs respectively. The novel MMPs display a morphological cell differentiation, implying a potential division of labour. These findings provide new insights into the diversity of MMPs in general, and contribute to our understanding of the evolution of multicellularity among prokaryotes.

Novel rod-shaped magnetotactic bacteria belonging to the class Alphaproteobacteria.

Novel large, rod-shaped magnetotactic bacteria (MTB) were discovered in intertidal sediments of the Yellow Sea, China. They biomineralized more than 300 rectangular magnetite magnetosomes per cell. Phylogenetic analysis based on the 16S rRNA gene sequence revealed that they are affiliated with the Alphaproteobacteria and may represent a new genus of MTB.

Slow N-acetyltransferase 2 genotype contributes to anti-tuberculosis drug-induced hepatotoxicity: a meta-analysis.

Pathogenesis and genetic factors influencing predisposition to antituberculosis drug-induced hepatotoxicity (ATDH) are not clear. Polymorphism at the genetic locus of a drug and xenobiotic compound metabolizing enzyme, N-acetyltransferase type 2 (NAT2), is reported to be associated with the excess generation of toxic reactive metabolites. To date, many case-control studies have been carried out to investigate the relationship between the NAT2 polymorphisms and ATDH, but the results have been inconsistent. To investigate this inconsistency, a meta-analysis was performed. Databases including PubMed, Web of Science, EMBASE and CNKI were searched to find relevant studies. A total of 26 case-control studies, involving 1,198 cases and 2,921 controls were included. Overall, we found significant association between slow acetylator genotype of NAT2 and ATDH (OR = 3.10, 95% CI: 2.47-3.88, P < 10(-5)). Significant results were also found in East Asians, South Asians, Brazilians and Middle Eastern when stratified by ethnicity. However, no significant associations were found for Caucasians. This meta-analysis demonstrated that the slow acetylator genotype of NAT2 is a risk factor associated with increased ATDH susceptibility, but these associations vary in different ethnic populations.

Magnetosome Gene Duplication as an Important Driver in the Evolution of Magnetotaxis in the Alphaproteobacteria.

The evolution of microbial magnetoreception (or magnetotaxis) is of great interest in the fields of microbiology, evolutionary biology, biophysics, geomicrobiology, and geochemistry. Current genomic data from magnetotactic bacteria (MTB), the only prokaryotes known to be capable of sensing the Earth's geomagnetic field, suggests an ancient origin of magnetotaxis in the domain Bacteria Vertical inheritance, followed by multiple independent magnetosome gene cluster loss, is considered to be one of the major forces that drove the evolution of magnetotaxis at or above the class or phylum level, although the evolutionary trajectories at lower taxonomic ranks (e.g., within the class level) remain largely unstudied. Here we report the isolation, cultivation, and sequencing of a novel magnetotactic spirillum belonging to the genus Terasakiella (Terasakiella sp. strain SH-1) within the class Alphaproteobacteria The complete genome sequence of Terasakiella sp. strain SH-1 revealed an unexpected duplication event of magnetosome genes within the mamAB operon, a group of genes essential for magnetosome biomineralization and magnetotaxis. Intriguingly, further comparative genomic analysis suggests that the duplication of mamAB genes is a common feature in the genomes of alphaproteobacterial MTB. Taken together, with the additional finding that gene duplication appears to have also occurred in some magnetotactic members of the Deltaproteobacteria, our results indicate that gene duplication plays an important role in the evolution of magnetotaxis in the Alphaproteobacteria and perhaps the domain Bacteria IMPORTANCE A diversity of organisms can sense the geomagnetic field for the purpose of navigation. Magnetotactic bacteria are the most primitive magnetism-sensing organisms known thus far and represent an excellent model system for the study of the origin, evolution, and mechanism of microbial magnetoreception (or magnetotaxis). The present study is the first report focused on magnetosome gene cluster duplication in the Alphaproteobacteria, which suggests the important role of gene duplication in the evolution of magnetotaxis in the Alphaproteobacteria and perhaps the domain Bacteria A novel scenario for the evolution of magnetotaxis in the Alphaproteobacteria is proposed and may provide new insights into evolution of magnetoreception of higher species.

Seasonal changes in the vertical distribution of two types of multicellular magnetotactic prokaryotes in the sediment of Lake Yuehu, China.

There are two genetically distinct morphological types of multicellular magnetotactic prokaryotes (MMPs) in the intertidal zone of Lake Yuehu (China): ellipsoidal MMPs (eMMPs) and spherical MMPs (sMMPs). We studied the vertical distribution of both types of MMPs in the sediment at Lake Yuehu during 1 year. Both types of MMPs were observed at sediment depths ranging from 1 to 34 cm, depending on the seasons. The eMMPs distributed at depths of 2-34 cm during spring, 1-11 cm during summer, 2-21 cm during autumn and 9-32 cm during winter. The eMMP species Candidatus Magnetananas rongchenensis, with magnetite magnetosomes, dominated at all distribution depths. These results suggested that Ca. M. rongchenensis migrated vertically during four seasons. The vertical profiles of oxidation-reduction potential (ORP) in Lake Yuehu changed seasonally, and these changes coincided with the seasonal distribution of MMPs, suggesting that the ORP affected the vertical distribution of MMPs. In addition, high concentrations of ammonium and silicate were associated with low abundances of MMPs.

Novel species and expanded distribution of ellipsoidal multicellular magnetotactic prokaryotes.

Multicellular magnetotactic prokaryotes (MMPs) are a peculiar group of magnetotactic bacteria, each comprising approximately 10-100 cells of the same phylotype. Two morphotypes of MMP have been identified, including several species of globally distributed spherical mulberry-like MMPs (s-MMPs), and two species of ellipsoidal pineapple-like MMPs (e-MMPs) from China (Qingdao and Rongcheng cities). We recently collected e-MMPs from Mediterranean Sea sediments (Six-Fours-les-Plages) and Drummond Island, in the South China Sea. Phylogenetic analysis revealed that the MMPs from Six-Fours-les-Plages and the previously reported e-MMP Candidatus Magnetananas rongchenensis have 98.5% sequence identity and are the same species, while the MMPs from Drummond Island appear to be a novel species, having > 7.1% sequence divergence from the most closely related e-MMP, Candidatus Magnetananas tsingtaoensis. Identification of the novel species expands the distribution of e-MMPs to Tropical Zone. Comparison of nine physical and chemical parameters revealed that sand grain size and the content of inorganic nitrogen (nitrate, ammonium and nitrite) in the sediments from Rongcheng City and Six-Fours-les-Plages were similar, and lower than found for sediments from the other two sampling sites. The results of the study reveal broad diversity and wide distribution of e-MMPs.

Temporal distributions and environmental adaptations of two types of multicellular magnetotactic prokaryote in the sediments of Lake Yuehu, China.

Two morphotypes (spherical and ellipsoidal) of multicellular magnetotactic prokaryotes (MMPs) have been reported from the sediments of Lake Yuehu, China. Here, their temporal distributions and their relationships with biogeochemical parameters are studied. Samples were collected at approximately 2-week intervals from two sites (A and B) during the period September 2012 to December 2013. The abundance of MMPs was high in summer and autumn, but low in winter and spring. Furthermore, the peaks in the numbers of the two types of MMPs were sequential, with the highest concentration of the spherical MMPs occurring prior to that of the ellipsoidal MMPs. This may be related to different optimal growth temperatures for the two types. Although the two types of MMP coexisted at both sites, their numbers were different; at most times, spherical MMPs dominated at site A, whereas ellipsoidal MMPs dominated at site B. Geochemical analysis revealed that the environmental conditions at site A varied more than at site B. Compared with the widely distributed spherical MMPs, ellipsoidal MMPs seemed to prefer more stable habitats. This is the first report of the temporal distribution of ellipsoidal MMPs in sediments, suggesting that their environmental adaptations differ from those of spherical MMPs.

Characterization and phylogenetic identification of a species of spherical multicellular magnetotactic prokaryotes that produces both magnetite and greigite crystals.

Magnetotactic bacteria (MTB) are a group of Gram-negative bacteria synthesizing magnetic crystals that allow them aligning along magnetic field lines. They have diverse morphologies including cocci, rods, vibrio, spirilla, and multicellular magnetotactic prokaryotes (MMPs). MMPs are composed of 10-57 cells with peritrichous flagella on their outer surfaces and swim as an entire unit. Here, we describe a species of spherical MMPs isolated from intertidal sediments of Lake Yuehu (Yellow Sea, China). They were mainly found in the subsurface layer of gray-black sediments. Microscopy revealed that these spherical MMPs were 5.6 ± 0.9 µm in diameter and composed of approximately 16-32 ovoid cells with a helical arrangement and peritrichous flagellation. High-resolution transmission electron microscopy showed that the MMPs contained both bullet-shaped magnetite and irregular greigite magnetosomes that were arranged in chains or clusters. These MMPs displayed typical escape motility and negative phototaxis. The 16S rRNA genes of micromanipulation-purified spherical MMPs were cloned and sequenced. Phylogenetic analysis revealed that the MMP species was affiliated with Deltaproteobacteria and displayed >2.8% sequence divergence with respect to previously reported MMPs. This is the first phylogenetic identification of a spherical MMP that produces both magnetite and greigite magnetosomes.