COVID-19 detection in chest X-ray images using deep boosted hybrid learning.

The new emerging COVID-19, declared a pandemic disease, has affected millions of human lives and caused a massive burden on healthcare centers. Therefore, a quick, accurate, and low-cost computer-based tool is required to timely detect and treat COVID-19 patients. In this work, two new deep learning frameworks: Deep Hybrid Learning (DHL) and Deep Boosted Hybrid Learning (DBHL), is proposed for effective COVID-19 detection in X-ray dataset. In the proposed DHL framework, the representation learning ability of the two developed COVID-RENet-1 & 2 models is exploited individually through a machine learning (ML) classifier. In COVID-RENet models, Region and Edge-based operations are carefully applied to learn region homogeneity and extract boundaries features. While in the case of the proposed DBHL framework, COVID-RENet-1 & 2 are fine-tuned using transfer learning on the chest X-rays. Furthermore, deep feature spaces are generated from the penultimate layers of the two models and then concatenated to get a single enriched boosted feature space. A conventional ML classifier exploits the enriched feature space to achieve better COVID-19 detection performance. The proposed COVID-19 detection frameworks are evaluated on radiologist's authenticated chest X-ray data, and their performance is compared with the well-established CNNs. It is observed through experiments that the proposed DBHL framework, which merges the two-deep CNN feature spaces, yields good performance (accuracy: 98.53%, sensitivity: 0.99, F-score: 0.98, and precision: 0.98). Furthermore, a web-based interface is developed, which takes only 5-10s to detect COVID-19 in each unseen chest X-ray image. This web-predictor is expected to help early diagnosis, save precious lives, and thus positively impact society.

Complete genome sequence of Staphylococcus aureus, strain ILRI_Eymole1/1, isolated from a Kenyan dromedary camel.

We report the genome of a Staphylococcus aureus strain (ILRI_Eymole1/1) isolated from a nasal swab of a dromedary camel (Camelus dromedarius) in North Kenya. The complete genome sequence of this strain consists of a circular chromosome of 2,874,302 bp with a GC-content of 32.88 %. In silico annotation predicted 2755 protein-encoding genes and 76 non-coding genes. This isolate belongs to MLST sequence type 30 (ST30). Phylogenetic analysis based on a subset of 283 core genes revealed that it falls within the human clonal complex 30 (CC30) S. aureus isolate cluster but is genetically distinct. About 79 % of the protein encoding genes are part of the CC30 core genome (genes common to all CC30 S. aureus isolates), ~18 % were within the variable genome (shared among multiple but not all isolates) and ~ 3 % were found only in the genome of the camel isolate. Among the 85 isolate-specific genes, 79 were located within putative phages and pathogenicity islands. Protein encoding genes associated with bacterial adhesion, and secretory proteins that are essential components of the type VII secretion system were also identified. The complete genome sequence of S. aureus strain ILRI_Eymole1/1 has been deposited in the European Nucleotide Archive under the accession no LN626917.1.

Brachyspira suanatina sp. nov., an enteropathogenic intestinal spirochaete isolated from pigs and mallards: genomic and phenotypic characteristics.

BACKGROUND: The genus Brachyspira currently encompasses seven valid species that colonize the intestines of mammals and birds. In a previous study a group of strongly haemolytic isolates from pigs and mallards was provisionally described as a new species within genus Brachyspira, "B. suanatina", and enteropathogenic properties were demonstrated in a porcine challenge model. METHODS: In the current study characterization of B. suanatina was performed on the basis of cell morphology, growth characteristics, enzyme profiles, DNA-DNA hybridization (DDH) and whole genome comparisons. The draft genome sequence of B. suanatina strain AN4859/03 was determined and compared with the available genomes of all valid species of Brachyspira. RESULTS: According to morphological traits, growth characteristics and enzymatic profiles, B. suanatina was similar to the type strain of B. hyodysenteriae, but using the recommended threshold value of 70% similarity by DDH it did not belong to any of the recognized Brachyspira species (range 16-64% similarity). This was further supported by average nucleotide identity values. Phylogenetic analysis performed using housekeeping genes and core genomes of all valid Brachyspira sp. and "B. hampsonii" revealed that B. suanatina and B. intermedia formed a clade distinct from B. hyodysenteriae. By comparing the genomes of the three closely related species B. intermedia, B. hyodysenteriae and B. suanatina similar profiles of general genomic features and distribution of genes in different functional categories were obtained. However, the genome size of B. hyodysenteriae was smallest among the species, suggesting the possibility of reductive evolution in the divergence of this species. A bacteriophage region and a putative plasmid sequence were also found in the genome of B. suanatina strain AN4859/03. CONCLUSIONS: The results of our study suggest that despite being similar to B. hyodysenteriae phenotypically, B. suanatina should be regarded as a separate species based on its genetic characteristics. Based on characteristics presented in this report we propose that strains AN4859/03, AN1681:1/04, AN2384/04 and Dk12570-2 from pigs in Sweden and Denmark, and strains AN3949:2/02 and AN1418:2/01 isolated from mallards in Sweden, represent a unique species within genus Brachyspira. For this new species we propose the name B. suanatina for which the type strain is AN4859/03T (=ATCC® BAA-2592™=DSM 100974T).

Genome Sequences of Two Pathogenic Streptococcus agalactiae Isolates from the One-Humped Camel Camelus dromedarius.

Streptococcus agalactiae causes a range of clinical syndromes in camels (Camelus dromedarius). We report the genome sequences of two S. agalactiae isolates that induce abscesses in Kenyan camels. These genomes provide novel data on the composition of the S. agalactiae "pan genome" and reveal the presence of multiple genomic islands.

Candidate genes prediction in pakistani families with hearing impairment by using bioinformatic approach / Predição de genes candidatos em famílias paquistanesas com deficiência auditiva utilizando uma abordagem de bioinformática

Multiple factors such as genetic and environmental, are involved in causing hearing impairment (HI). Severe or profound hearing loss affects approximately one in 1000 children worldwide and half of these cases are due to genetic factors. In case of hereditary nonsyndromic HI, approximately 75­80% of cases are involved in autosomal recessive inheritance and 15% of cases involve autosomal dominant inheritance. HI represents extreme genetic heterogeneity. In nonsyndromic deafness, 135 loci have been mapped till now including 77 autosomal recessive genes of which only 29 corresponding nuclear genes have been cloned. This study was designed to apply bioinformatic approach for reducing large number of candidate genes responsible for deafness to a handy number for their mutation analysis. Databases of expressed mouse inner ear genes and the expressed human cochlear genes were used to cross-reference all genes present in particular locus predicting candidate genes for phenotypes of nonsyndromic hereditary HI. These candidate genes are a source of starting point for mutation analysis along with genetic linkage to refine the loci. After characterization, it was observed that KIAA119 and EDN3 are candidate genes for deafness. In present study, there were total 14 loci and two genes KIAA119 and EDN3 were identified as candidate genes in locus 48 and locus 65 respectively. If mutation analysis of the two characterized genes is done, it will not be a comparatively time taking and labor-intensive process as these genes are only two in number.

Diversos fatores, tais como genéticos e ambientais, estão envolvidos na causa da deficiência auditiva (HI). A perda auditiva severa ou profunda afeta aproximadamente uma em cada 1000 crianças em todo o mundo e metade destes casos são devidos a fatores genéticos. Em relação a HI não-sindrômica hereditária, cerca de 75-80% dos casos estão envolvidos na herança autossômica recessiva e 15% dos casos envolvem herança autossômica dominante. HI representa extrema heterogeneidade genética. Em casos de surdez, 135 loci foram mapeados até agora, incluindo 77 genes autossômicos recessivos das quais apenas 29 genes correspondentes nucleares foram clonados. Este estudo foi desenhado para aplicar abordagem de bioinformática a fim de reduzir o grande número de genes candidatos responsáveis pela surdez a um número útil para a análise de mutação. Bases de dados de genes expressos do ouvido interno em camundongos e de genes expressos na cóclea em humanos foram usados para cruzar todos os genes presentes no locus específico prevendo genes candidatos para os fenótipos de HI não sindrômica hereditária. Estes genes candidatos são uma fonte de ponto de partida para a análise de mutação, juntamente com a ligação gênica para refinar os locos. Após a caracterização, verificouse que KIAA119 e EDN3 são genes candidatos para a surdez. No presente estudo, houve um total de 14 locos e dois genes KIAA119 e EDN3 foram identificados como genes candidatos no locus 48 e locus 65, respectivamente. Se a análise de mutação dos dois genes caracterizados for feita, não será um processo comparativamente longo e trabalhoso uma vez que são apenas dois genes.