A face image classification method of autistic children based on the two-phase transfer learning.

Autism spectrum disorder (ASD) is a neurodevelopmental disorder, which seriously affects children's normal life. Screening potential autistic children before professional diagnose is helpful to early detection and early intervention. Autistic children have some different facial features from non-autistic children, so the potential autistic children can be screened by taking children's facial images and analyzing them with a mobile phone. The area under curve (AUC) is a more robust metrics than accuracy in evaluating the performance of a model used to carry out the two-category classification, and the AUC of the deep learning model suitable for the mobile terminal in the existing research can be further improved. Moreover, the size of an input image is large, which is not fit for a mobile phone. A deep transfer learning method is proposed in this research, which can use images with smaller size and improve the AUC of existing studies. The proposed transfer method uses the two-phase transfer learning mode and the multi-classifier integration mode. For MobileNetV2 and MobileNetV3-Large that are suitable for a mobile phone, the two-phase transfer learning mode is used to improve their classification performance, and then the multi-classifier integration mode is used to integrate them to further improve the classification performance. A multi-classifier integrating calculation method is also proposed to calculate the final classification results according to the classifying results of the participating models. The experimental results show that compared with the one-phase transfer learning, the two-phase transfer learning can significantly improve the classification performance of MobileNetV2 and MobileNetV3-Large, and the classification performance of the integrated classifier is better than that of any participating classifiers. The accuracy of the integrated classifier in this research is 90.5%, and the AUC is 96.32%, which is 3.51% greater than the AUC (92.81%) of the previous studies.

Isolation and Genomics of Futiania mangrovii gen. nov., sp. nov., a Rare and Metabolically Versatile Member in the Class Alphaproteobacteria.

Mangrove microorganisms are a major part of the coastal ecosystem and are directly associated with nutrient cycling. Despite their ecological significance, the collection of culturable mangrove microbes is limited due to difficulties in isolation and cultivation. Here, we report the isolation and genome sequence of strain FT118T, the first cultured representative of a previously uncultivated order UBA8317 within Alphaproteobacteria, based on the combined results of 16S rRNA gene similarity, phylogenomic, and average amino acid identity analyses. We propose Futianiales ord. nov. and Futianiaceae fam. nov. with Futiania as the type genus, and FT118T represents the type species with the name Futiania mangrovii gen. nov, sp. nov. The 16S rRNA gene sequence comparison reveals that this novel order is a rare member but has a ubiquitous distribution across various habitats worldwide, which is corroborated by the experimental confirmation that this isolate can physiologically adapt to a wide range of oxygen levels, temperatures, pH and salinity levels. Biochemical characterization, genomic annotation, and metatranscriptomic analysis of FT118T demonstrate that it is metabolically versatile and active in situ. Genomic analysis reveals adaptive features of Futianiales to fluctuating mangrove environments, including the presence of high- and low-affinity terminal oxidases, N-type ATPase, and the genomic capability of producing various compatible solutes and polyhydroxybutyrate, which possibly allow for the persistence of this novel order across various habitats. Collectively, these results expand the current culture collection of mangrove microorganisms, providing genomic insights of how this novel taxon adapts to fluctuating environments and the culture reference to unravel possible microbe-environment interactions. IMPORTANCE The rare biosphere constitutes an essential part of the microbial community and may drive nutrient cycling and other geochemical processes. However, the difficulty in microbial isolation and cultivation has hampered our understanding of the physiology and ecology of uncultured rare lineages. In this study, we successfully isolated a novel alphaproteobacterium, designated as FT118T, and performed a combination of phenotypic, phylogenetic, and phylogenomic analyses, confirming that this isolate represents the first cultured member of a previously uncultivated order UBA8317 within Alphaproteobacteria. It is a rare species with a ubiquitous distribution across different habitats. Genomic and metatranscriptomic analyses demonstrate that it is metabolically versatile and active in situ, suggesting its potential role in nutrient cycling despite being scarce. This work not only expands the current phylogeny of isolated Alphaproteobacteria but also provides genomic and culture reference to unravel microbial adaptation strategies in mangrove sediments and possible microbe-environment interactions.

Comparative genomic analysis reveals metabolic flexibility of Woesearchaeota.

The archaeal phylum Woesearchaeota, within the DPANN superphylum, includes phylogenetically diverse microorganisms that inhabit various environments. Their biology is poorly understood due to the lack of cultured isolates. Here, we analyze datasets of Woesearchaeota 16S rRNA gene sequences and metagenome-assembled genomes to infer global distribution patterns, ecological preferences and metabolic capabilities. Phylogenomic analyses indicate that the phylum can be classified into ten subgroups, termed A-J. While a symbiotic lifestyle is predicted for most, some members of subgroup J might be host-independent. The genomes of several Woesearchaeota, including subgroup J, encode putative [FeFe] hydrogenases (known to be important for fermentation in other organisms), suggesting that these archaea might be anaerobic fermentative heterotrophs.

Expanded diversity of Asgard archaea and their relationships with eukaryotes.

Asgard is a recently discovered superphylum of archaea that appears to include the closest archaeal relatives of eukaryotes1-5. Debate continues as to whether the archaeal ancestor of eukaryotes belongs within the Asgard superphylum or whether this ancestor is a sister group to all other archaea (that is, a two-domain versus a three-domain tree of life)6-8. Here we present a comparative analysis of 162 complete or nearly complete genomes of Asgard archaea, including 75 metagenome-assembled genomes that-to our knowledge-have not previously been reported. Our results substantially expand the phylogenetic diversity of Asgard and lead us to propose six additional phyla that include a deep branch that we have provisionally named Wukongarchaeota. Our phylogenomic analysis does not resolve unequivocally the evolutionary relationship between eukaryotes and Asgard archaea, but instead-depending on the choice of species and conserved genes used to build the phylogeny-supports either the origin of eukaryotes from within Asgard (as a sister group to the expanded Heimdallarchaeota-Wukongarchaeota branch) or a deeper branch for the eukaryote ancestor within archaea. Our comprehensive protein domain analysis using the 162 Asgard genomes results in a major expansion of the set of eukaryotic signature proteins. The Asgard eukaryotic signature proteins show variable phyletic distributions and domain architectures, which is suggestive of dynamic evolution through horizontal gene transfer, gene loss, gene duplication and domain shuffling. The phylogenomics of the Asgard archaea points to the accumulation of the components of the mobile archaeal 'eukaryome' in the archaeal ancestor of eukaryotes (within or outside Asgard) through extensive horizontal gene transfer.

Ecological features and global distribution of Asgard archaea.

Asgard is a newly proposed archaeal superphylum, which has been suggested to hold the key to decipher the origin of Eukaryotes. However, their ecology remains largely unknown. Here, we conducted a meta-analysis of publicly available Asgard-associated 16S rRNA gene fragments, and found that just three previously proposed clades (Lokiarchaeota, Thorarchaeota, and Asgard clade 4) are widely distributed, whereas the other seven clades (phylum or class level) are restricted to the sediment biosphere. Asgard archaea, especially Loki- and Thorarchaeota, seem to adapt to marine sediments, and water depth (the depth of the sediment below water surface) and salinity might be crucial factors for the proportion of these microorganisms as revealed by multivariate regression analyses. However, the abundance of Asgard archaea exhibited distinct environmental drivers at the clade-level; for instance, the proportion of Asgard clade 4 was higher in less saline environments (salinity <6.35 psu), while higher for Heimdallarchaeota-AAG and Asgard clade 2 in more saline environment (salinity ≥35 psu). Furthermore, co-occurrence analysis allowed us to find a significant non-random association of different Asgard clades with other groups (e.g., Lokiarchaeota with Deltaproteobacteria and Anaerolineae; Odinarchaeota with Bathyarchaeota), suggesting different interaction potentials among these clades. Overall, these findings reveal Asgard archaea as a ubiquitous group worldwide and provide initial insights into their ecological features on a global scale.

Comparative genomic analysis reveals metabolic diversity of different Paenibacillus groups.

The genus Paenibacillus was originally recognized based on the 16S rRNA gene phylogeny. Recently, a standardized bacterial taxonomy approach based on a genome phylogeny has substantially revised the classification of Paenibacillus, dividing it into 23 genera. However, the metabolic differences among these groups remain undescribed. Here, genomes of 41 Paenibacillus strains comprising 25 species were sequenced, and a comparative genomic analysis was performed considering these and 187 publicly available Paenibacillus genomes to understand their phylogeny and metabolic differences. Phylogenetic analysis indicated that Paenibacillus clustered into 10 subgroups. Core genome and pan-genome analyses revealed similar functional categories among the different Paenibacillus subgroups; however, each group tended to harbor specific gene families. A large proportion of genes in the subgroups A, E, and G are related to carbohydrate metabolism. Among them, genes related to the glycoside hydrolase family were most abundant. Metabolic reconstruction of the newly sequenced genomes showed that the Embden-Meyerhof-Parnas pathway, pentose phosphate pathway, and citric acid cycle are central pathways of carbohydrate metabolism in Paenibacillus. Further, the genomes of the subgroups A and G lack genes involved in glyoxylate cycle and D-galacturonate degradation, respectively. The current study revealed the metabolic diversity of Paenibacillus subgroups assigned based on a genomic phylogeny and could inform the taxonomy of Paenibacillus. KEY POINTS: • Paenibacillus clustered into 10 subgroups. • Genomic content variation and metabolic diversity in the subgroup A, E, and G were described. • Carbohydrate transport and metabolism is important for Paenibacillus survival.

Dynamics of microbial community in the bioreactor for bisphenol S removal.

Bisphenol S is one of the alternative substitutes of Bisphenol A, a chemical widely recognized as an endocrine disrupting compound. In the past few years, a variety of studies on degradation of BPA demonstrated that microorganisms play important roles in the degradation process. However, the fate of BPS during wastewater treatment processes and the composition of microorganisms that functionalize BPS degradation remain to be explored. In this study, three bioreactors, R-BPS (amended with Bisphenol S), R-BPSHA (amended with Bisphenol S and humic acid) and Con (control bioreactor), were set up to investigate the fate of BPS and the microbial compositions and dynamics in the bioreactors, especially for the microorganisms associated with BPS removal. Results showed that a complete removal was achieved within 24 days. The addition of humic acid accelerated the elimination of BPS in both effluent and sludge. The results of 16S rRNA gene ampilicon sequencing revealed that the most abundant bacteria in all samples were affiliated to Proteobacteria, Bacteroidetes, Acidobacteria and Chloroflexi. Seven major genera were likely associated with BPS removal, including Pseudomonas, Azospira, Hydrogenophaga, Devosia, Delftia, Acidovorax and Rhodobacter. Among them, humic acid increased relative abundance of some bacteria, such as Pseudomonas, Hydrogenophaga and Acidovorax. These findings would give valuable information on the microbial community composition associated with BPS removal, providing biological background for bioremediation of BPS-contaminated environment.