Sequence-based Classification Research Articles

Genome sequence-based species classification of Enterobacter cloacae complex: a study among clinical isolates.

Standardizing taxonomy of the Enterobacter cloacae complex (ECC) is necessary for data integration across diverse studies. The study utilized whole-genome data to accurately identify 256 clinical ECC isolated from bloodstream infections using average nucleotide identity (ANI), digitalized DNA-DNA hybridization (dDDH), and phylogenetic analysis. Through comprehensive assessments including phylogenetic analysis, ANI and dDDH comparisons, virulence gene, and capsule typing of the 256 clinical isolates, it was concluded that the classification method based on subspecies exhibited better correlation and consistency with the molecular characteristics of clinical ECC isolates. In summary, this research contributes to the precise identification of clinical ECC at the species level and expands our understanding of ECC.

Unlocking the genome of the non-sourdough Kazachstania humilis MAW1: insights into inhibitory factors and phenotypic properties

BackgroundAscomycetous budding yeasts are ubiquitous environmental microorganisms important in food production and medicine. Due to recent intensive genomic research, the taxonomy of yeast is becoming more organized based on the identification of monophyletic taxa. This includes genera important to humans, such as Kazachstania. Until now, Kazachstania humilis (previously Candida humilis) was regarded as a sourdough-specific yeast. In addition, any antibacterial activity has not been associated with this species.ResultsPreviously, we isolated a yeast strain that impaired bio-hydrogen production in a dark fermentation bioreactor and inhibited the growth of Gram-positive and Gram-negative bacteria. Here, using next generation sequencing technologies, we sequenced the genome of this strain named K. humilis MAW1. This is the first genome of a K. humilis isolate not originating from a fermented food. We used novel phylogenetic approach employing the 18 S-ITS-D1-D2 region to show the placement of the K. humilis MAW1 among other members of the Kazachstania genus. This strain was examined by global phenotypic profiling, including carbon sources utilized and the influence of stress conditions on growth. Using the well-recognized bacterial model Escherichia coli AB1157, we show that K. humilis MAW1 cultivated in an acidic medium inhibits bacterial growth by the disturbance of cell division, manifested by filament formation. To gain a greater understanding of the inhibitory effect of K. humilis MAW1, we selected 23 yeast proteins with recognized toxic activity against bacteria and used them for Blast searches of the K. humilis MAW1 genome assembly. The resulting panel of genes present in the K. humilis MAW1 genome included those encoding the 1,3-β-glucan glycosidase and the 1,3-β-glucan synthesis inhibitor that might disturb the bacterial cell envelope structures.ConclusionsWe characterized a non-sourdough-derived strain of K. humilis, including its genome sequence and physiological aspects. The MAW1, together with other K. humilis strains, shows the new organization of the mating-type locus. The revealed here pH-dependent ability to inhibit bacterial growth has not been previously recognized in this species. Our study contributes to the building of genome sequence-based classification systems; better understanding of K.humilis as a cell factory in fermentation processes and exploring bacteria-yeast interactions in microbial communities.

Detection and classification of the integrative conjugative elements of Lactococcus lactis

Lactococcus lactis is widely applied by the dairy industry for the fermentation of milk into products such as cheese. Adaptation of L. lactis to the dairy environment often depends on functions encoded by mobile genetic elements (MGEs) such as plasmids. Other L. lactis MGEs that contribute to industrially relevant traits like antimicrobial production and carbohydrate utilization capacities belong to the integrative conjugative elements (ICE). Here we investigate the prevalence of ICEs in L. lactis using an automated search engine that detects colocalized, ICE-associated core-functions (involved in conjugation or mobilization) in lactococcal genomes. This approach enabled the detection of 36 candidate-ICEs in 69 L. lactis genomes. By phylogenetic analysis of conserved protein functions encoded in all lactococcal ICEs, these 36 ICEs could be classified in three main ICE-families that encompass 7 distinguishable ICE-integrases and are characterized by apparent modular-exchangeability and plasticity. Finally, we demonstrate that phylogenetic analysis of the conjugation-associated VirB4 ATPase function differentiates ICE- and plasmid-derived conjugation systems, indicating that conjugal transfer of lactococcal ICEs and plasmids involves genetically distinct machineries. Our genomic analysis and sequence-based classification of lactococcal ICEs creates a comprehensive overview of the conserved functional repertoires encoded by this family of MGEs in L. lactis, which can facilitate the future exploitation of the functional traits they encode by ICE mobilization to appropriate starter culture strains.

Palm Fungi and Their Key Role in Biodiversity Surveys: A Review.

Over the past three decades, a wealth of studies has shown that palm trees (Arecaceae) are a diverse habitat with intense fungal colonisation, making them an important substratum to explore fungal diversity. Palm trees are perennial, monocotyledonous plants mainly restricted to the tropics that include economically important crops and highly valued ornamental plants worldwide. The extensive research conducted in Southeast Asia and Australasia indicates that palm fungi are undoubtedly a taxonomically diverse assemblage from which a remarkable number of new species is continuously being reported. Despite this wealth of data, no recent comprehensive review on palm fungi exists to date. In this regard, we present here a historical account and discussion of the research on the palm fungi to reflect on their importance as a diverse and understudied assemblage. The taxonomic structure of palm fungi is also outlined, along with comments on the need for further studies to place them within modern DNA sequence-based classifications. Palm trees can be considered model plants for studying fungal biodiversity and, therefore, the key role of palm fungi in biodiversity surveys is discussed. The close association and intrinsic relationship between palm hosts and palm fungi, coupled with a high fungal diversity, suggest that the diversity of palm fungi is still far from being fully understood. The figures suggested in the literature for the diversity of palm fungi have been revisited and updated here. As a result, it is estimated that there are about 76,000 species of palm fungi worldwide, of which more than 2500 are currently known. This review emphasises that research on palm fungi may provide answers to a number of current fungal biodiversity challenges.

Genomic epidemiology and heterogeneity of Providencia and their bla NDM-1-carrying plasmids

ABSTRACT Providencia as an opportunistic pathogen can cause serious infection, and moreover the emergence of multi-drug-resistant Providencia strains poses a potentially life-threatening risk to public health. However, a comprehensive genomic study to reveal the population structure and dissemination of Providencia is still lacking. In this study, we conducted a genomic epidemiology analysis on the 580 global sequenced Providencia isolates, including 257 ones sequenced in this study (42 ones were fully sequenced). We established a genome sequence-based species classification scheme for Providencia, redefining the conventional 11 Providencia species into seven genocomplexes that were further divided into 18 genospecies, providing an extensively updated reference for Providencia species discrimination based on the largest Providencia genome dataset to date. We then dissected the profile of antimicrobial resistance genes and the prevalence of multi-drug-resistant Providencia strains among these genocomplexes/genospecies, disclosing the presence of diverse and abundant antimicrobial resistance genes and high resistance ratios against multiple classes of drugs in Providencia. We further dissected the genetic basis for the spread of bla NDM-1 in Providencia. bla NDM-1 genes were mainly carried by five incompatible (Inc) groups of plasmids: IncC, IncW, IncpPROV114-NR, IncpCHS4.1-3, and IncpPrY2001, and the last three were newly designated in this study. By tracking the spread of bla NDM-1-carrying plasmids, IncC, IncpPROV114-NR, IncpCHS4.1-3, and IncpPrY2001 plasmids were found to be highly involved in parallel horizontal transfer or vertical clonal expansion of bla NDM-1 among Providencia. Overall, our study provided a comprehensive genomic view of species differentiation, antimicrobial resistance prevalence, and plasmid-mediated bla NDM-1 dissemination in Providencia.

The ESTHER database on alpha/beta hydrolase fold proteins - An overview of recent developments

The ESTHER database, dedicated to ESTerases and alpha/beta-Hydrolase Enzymes and Relatives (https://bioweb.supagro.inra.fr/ESTHER/general?what=index), offers online access to a continuously updated, sequence-based classification of proteins harboring the alpha/beta hydrolase fold into families and subfamilies. In particular, the database proposes links to the sequences, structures, ligands and huge diversity of functions of these proteins, and to the related literature and other databases. Taking advantage of the promiscuity of enzymatic function, many engineered esterases, lipases, epoxide-hydrolases, haloalkane dehalogenases are used for biotechnological applications. Finding means for detoxifying those protein members that are targeted by insecticides, herbicides, antibiotics, or for reactivating human cholinesterases when inhibited by nerve gas, are still active areas of research. Using or improving the capacity of some enzymes to breakdown plastics with the aim to recycle valuable material and reduce waste is an emerging challenge. Most hydrolases in the superfamily are water-soluble and act on or are inhibited by small organic compounds, yet in a few subfamilies some members interact with other, unrelated proteins to modulate activity or trigger functional partnerships. Recent development in 3D structure prediction brought by AI-based programs now permits analysis of enzymatic mechanisms for a variety of hydrolases with no experimental 3D structure available. Finally, mutations in as many as 34 of the 120 human genes compiled in the database are now linked to genetic diseases, a feature fueling research on early detection, metabolic pathways, pharmacological treatment or enzyme replacement therapy. Here we review those developments in the database that took place over the latest decade and discuss potential new applications and recent and future expected research in the field.

Genome-Wide Analysis of Haemonchus contortus Proteases and Protease Inhibitors Using Advanced Informatics Provides Insights into Parasite Biology and Host-Parasite Interactions.

Biodiversity within the animal kingdom is associated with extensive molecular diversity. The expansion of genomic, transcriptomic and proteomic data sets for invertebrate groups and species with unique biological traits necessitates reliable in silico tools for the accurate identification and annotation of molecules and molecular groups. However, conventional tools are inadequate for lesser-known organismal groups, such as eukaryotic pathogens (parasites), so that improved approaches are urgently needed. Here, we established a combined sequence- and structure-based workflow system to harness well-curated publicly available data sets and resources to identify, classify and annotate proteases and protease inhibitors of a highly pathogenic parasitic roundworm (nematode) of global relevance, called Haemonchus contortus (barber's pole worm). This workflow performed markedly better than conventional, sequence-based classification and annotation alone and allowed the first genome-wide characterisation of protease and protease inhibitor genes and gene products in this worm. In total, we identified 790 genes encoding 860 proteases and protease inhibitors representing 83 gene families. The proteins inferred included 280 metallo-, 145 cysteine, 142 serine, 121 aspartic and 81 "mixed" proteases as well as 91 protease inhibitors, all of which had marked physicochemical diversity and inferred involvements in >400 biological processes or pathways. A detailed investigation revealed a remarkable expansion of some protease or inhibitor gene families, which are likely linked to parasitism (e.g., host-parasite interactions, immunomodulation and blood-feeding) and exhibit stage- or sex-specific transcription profiles. This investigation provides a solid foundation for detailed explorations of the structures and functions of proteases and protease inhibitors of H. contortus and related nematodes, and it could assist in the discovery of new drug or vaccine targets against infections or diseases.

Machine Learning Based Prediction of Enzymatic Degradation of Plastics Using Encoded Protein Sequence and Effective Feature Representation

Enzyme biocatalysis for plastic treatment and recycling is an emerging field of growing interest. However, it is challenging and time-consuming to identify plastic-degrading enzymes with desirable functionality, given the large number of putative enzyme sequences. There is a critical need to develop an effective approach to accurately predict the enzyme activity in degrading different types of plastics. In this study, we developed a machine-learning-based plastic enzymatic degradation (PED) framework to predict the ability of an enzyme to degrade plastics of interest by exploring and recognizing hidden patterns in protein sequences. A data set integrating information from a wide range of experimentally verified enzymes and various common plastic substrates was created. A new context-aware enzyme sequence representation (CESR) mechanism was developed to learn the abundant contextual information in enzyme sequences, and feature extraction was performed for enzymes at both the amino acid level and global sequence level. Thirteen machine learning classification algorithms were compared, and XGBoost was identified as the best-performing algorithm. PED achieved an overall accuracy of 90.2% and outperformed sequence-based protein classification models from the existing literature. Furthermore, important enzyme features in plastic degradation were identified and comprehensively interpreted. This study demonstrated a new tool for the prediction and discovery of plastic-degrading enzymes.

Ion-pumping microbial rhodopsin protein classification by machine learning approach

BackgroundRhodopsin is a seven-transmembrane protein covalently linked with retinal chromophore that absorbs photons for energy conversion and intracellular signaling in eukaryotes, bacteria, and archaea. Haloarchaeal rhodopsins are Type-I microbial rhodopsin that elicits various light-driven functions like proton pumping, chloride pumping and Phototaxis behaviour. The industrial application of Ion-pumping Haloarchaeal rhodopsins is limited by the lack of full-length rhodopsin sequence-based classifications, which play an important role in Ion-pumping activity. The well-studied Haloarchaeal rhodopsin is a proton-pumping bacteriorhodopsin that shows promising applications in optogenetics, biosensitized solar cells, security ink, data storage, artificial retinal implant and biohydrogen generation. As a result, a low-cost computational approach is required to identify Ion-pumping Haloarchaeal rhodopsin sequences and its subtype.ResultsThis study uses a support vector machine (SVM) technique to identify these ion-pumping Haloarchaeal rhodopsin proteins. The haloarchaeal ion pumping rhodopsins viz., bacteriorhodopsin, halorhodopsin, xanthorhodopsin, sensoryrhodopsin and marine prokaryotic Ion-pumping rhodopsins like actinorhodopsin, proteorhodopsin have been utilized to develop the methods that accurately identified the ion pumping haloarchaeal and other type I microbial rhodopsins. We achieved overall maximum accuracy of 97.78%, 97.84% and 97.60%, respectively, for amino acid composition, dipeptide composition and hybrid approach on tenfold cross validation using SVM. Predictive models for each class of rhodopsin performed equally well on an independent data set. In addition to this, similar results were achieved using another machine learning technique namely random forest. Simultaneously predictive models performed equally well during five-fold cross validation. Apart from this study, we also tested the own, blank, BLAST dataset and annotated whole-genome rhodopsin sequences of PWS haloarchaeal isolates in the developed methods. The developed web server (https://bioinfo.imtech.res.in/servers/rhodopred) can identify the Ion Pumping Haloarchaeal rhodopsin proteins and their subtypes. We expect this web tool would be useful for rhodopsin researchers.ConclusionThe overall performance of the developed method results show that it accurately identifies the Ionpumping Haloarchaeal rhodopsin and their subtypes using known and unknown microbial rhodopsin sequences. We expect that this study would be useful for optogenetics, molecular biologists and rhodopsin researchers.

SulfAtlas, the sulfatase database: state of the art and new developments.

SulfAtlas (https://sulfatlas.sb-roscoff.fr/) is a knowledge-based resource dedicated to a sequence-based classification of sulfatases. Currently four sulfatase families exist (S1-S4) and the largest family (S1, formylglycine-dependent sulfatases) is divided into subfamilies by a phylogenetic approach, each subfamily corresponding to either a single characterized specificity (or few specificities in some cases) or to unknown substrates. Sequences are linked to their biochemical and structural information according to an expert scrutiny of the available literature. Database browsing was initially made possible both through a keyword search engine and a specific sequence similarity (BLAST) server. In this article, we will briefly summarize the experimental progresses in the sulfatase field in the last 6 years. To improve and speed up the (sub)family assignment of sulfatases in (meta)genomic data, we have developed a new, freely-accessible search engine using Hidden Markov model (HMM) for each (sub)family. This new tool (SulfAtlas HMM) is also a key part of the internal pipeline used to regularly update the database. SulfAtlas resource has indeed significantly grown since its creation in 2016, from 4550 sequences to 162 430 sequences in August 2022.

Nationwide and long-term molecular epidemiologic studies of mumps viruses that circulated in Japan between 1986 and 2017.

In Japan, major mumps outbreaks still occur every 4-5 years because of low mumps vaccine coverage (30-40%) owing to the voluntary immunization program. Herein, to prepare for a regular immunization program, we aimed to reveal the nationwide and long-term molecular epidemiological trends of the mumps virus (MuV) in Japan. Additionally, we performed whole-genome sequencing (WGS) using next-generation sequencing to assess results from conventional genotyping using MuV sequences of the small-hydrophobic (SH) gene. We analyzed 1,064 SH gene sequences from mumps clinical samples and MuV isolates collected from 25 prefectures from 1986 to 2017. The results showed that six genotypes, namely B (110), F (1), G (900), H (3), J (41), and L (9) were identified, and the dominant genotypes changed every decade in Japan since the 1980s. Genotype G has been exclusively circulating since the early 2000s. Seven clades were identified for genotype G using SH sequence-based classification. To verify the results, we performed WGS on 77 representative isolates of genotype G using NGS and phylogenetically analyzed them. Five clades were identified with high bootstrap values and designated as Japanese clade (JPC)-1, -2, -3, -4, -5. JPC-1 and -3 accounted for over 80% of the total genotype G isolates (68.3 and 13.8%, respectively). Of these, JPC-2 and -5, were newly identified clades in Japan through this study. This is the first report describing the nationwide and long-term molecular epidemiology of MuV in Japan. The results provide information about Japanese domestic genotypes, which is essential for evaluating the mumps elimination progress in Japan after the forthcoming introduction of the mumps vaccine into Japan's regular immunization program. Furthermore, the study shows that WGS analysis using NGS is more accurate than results obtained from conventional SH sequence-based classification and is a powerful tool for accurate molecular epidemiology studies.

Benchmarking online sequence-to-sequence and character-based handwriting recognition from IMU-enhanced pens

Handwriting is one of the most frequently occurring patterns in everyday life and with it comes challenging applications such as handwriting recognition, writer identification and signature verification. In contrast to offline HWR that only uses spatial information (i.e., images), online HWR uses richer spatio-temporal information (i.e., trajectory data or inertial data). While there exist many offline HWR datasets, there are only little data available for the development of OnHWR methods on paper as it requires hardware-integrated pens. This paper presents data and benchmark models for real-time sequence-to-sequence learning and single character-based recognition. Our data are recorded by a sensor-enhanced ballpoint pen, yielding sensor data streams from triaxial accelerometers, a gyroscope, a magnetometer and a force sensor at 100 Hz. We propose a variety of datasets including equations and words for both the writer-dependent and writer-independent tasks. Our datasets allow a comparison between classical OnHWR on tablets and on paper with sensor-enhanced pens. We provide an evaluation benchmark for seq2seq and single character-based HWR using recurrent and temporal convolutional networks and transformers combined with a connectionist temporal classification (CTC) loss and cross-entropy (CE) losses. Our convolutional network combined with BiLSTMs outperforms transformer-based architectures, is on par with InceptionTime for sequence-based classification tasks and yields better results compared to 28 state-of-the-art techniques. Time-series augmentation methods improve the sequence-based task, and we show that CE variants can improve the single classification task. Our implementations together with the large benchmark of state-of-the-art techniques of novel OnHWR datasets serve as a baseline for future research in the area of OnHWR on paper.

Naming the unnamed: over 65,000 Candidatus names for unnamed Archaea and Bacteria in the Genome Taxonomy Database.

Thousands of new bacterial and archaeal species and higher-level taxa are discovered each year through the analysis of genomes and metagenomes. The Genome Taxonomy Database (GTDB) provides hierarchical sequence-based descriptions and classifications for new and as-yet-unnamed taxa. However, bacterial nomenclature, as currently configured, cannot keep up with the need for new well-formed names. Instead, microbiologists have been forced to use hard-to-remember alphanumeric placeholder labels. Here, we exploit an approach to the generation of well-formed arbitrary Latinate names at a scale sufficient to name tens of thousands of unnamed taxa within GTDB. These newly created names represent an important resource for the microbiology community, facilitating communication between bioinformaticians, microbiologists and taxonomists, while populating the emerging landscape of microbial taxonomic and functional discovery with accessible and memorable linguistic labels.

Inferring Species Compositions of Complex Fungal Communities from Long- and Short-Read Sequence Data.

ABSTRACTThe kingdom Fungi is highly diverse in morphology and ecosystem function. Yet fungi are challenging to characterize as they can be difficult to culture and morphologically indistinct. Overall, their description and analysis lag far behind other microbes such as bacteria. Classification of species via high-throughput sequencing is increasingly becoming the norm for pathogen detection, microbiome studies, and environmental monitoring. With the rapid development of sequencing technologies, however, standardized procedures for taxonomic assignment of long sequence reads have not yet been well established. Focusing on nanopore sequencing technology, we compared classification and community composition analysis pipelines using shotgun and amplicon sequencing data generated from mock communities comprising 43 fungal species. We show that regardless of the sequencing methodology used, the highest accuracy of species identification was achieved by sequence alignment against a fungal-specific database. During the assessment of classification algorithms, we found that applying cutoffs to the query coverage of each read or contig significantly improved the classification accuracy and community composition analysis without major data loss. We also generated draft genome assemblies for three fungal species from nanopore data which were absent from genome databases. Our study improves sequence-based classification and estimation of relative sequence abundance using real fungal community data and provides a practical guide for the design of metagenomics analyses focusing on fungi.

A Genomic and Bioinformatics View of the Classification and Evolution of Morganella Species and Their Chromosomal Accessory Genetic Elements Harboring Antimicrobial Resistance Genes.

ABSTRACTIn this study, draft-genome sequencing was conducted for 60 Chinese Morganella isolates, and furthermore, 12 of them were fully sequenced. Then, a total of 166 global sequenced Morganella isolates, including the above 60, were collected to perform average nucleotide identity-based genomic classification and core single nucleotide polymorphism-based phylogenomic analysis. A genome sequence-based species classification scheme for Morganella was established, and accordingly, the two conventional Morganella species were redefined as two complexes and further divided into four and two genospecies, respectively. At least 88 acquired antimicrobial resistance genes (ARGs) were disseminated in these 166 isolates and were prevalent mostly in the isolates from hospital settings. IS26/IS15DI, IS10 and IS1R, and Tn3-, Tn21-, and Tn7-subfamily unit transposons were frequently presented in these 166 isolates. Furthermore, a detailed sequence comparison was applied to 18 Morganella chromosomal accessory genetic elements (AGEs) from the fully sequenced 12 isolates, together with 5 prototype AGEs from GenBank. These 23 AGEs were divided into eight different groups belonging to composite/unit transposons, transposable prophages, integrative and mobilizable elements, and integrative and conjugative elements, and they harbored at least 52 ARGs involved in resistance to 15 categories of antimicrobials. Eleven of these 23 AGEs acquired large accessory modules, which exhibited complex mosaic structures and contained many antimicrobial resistance loci and associated ARGs. Integration of ARG-containing AGEs into Morganella chromosomes would contribute to the accumulation and dissemination of ARGs in Morganella and enhance the adaption and survival of Morganella under complex and diverse antimicrobial selection pressures.IMPORTANCE This study presents a comprehensive genomic epidemiology analysis on global sequenced Morganella isolates. First, a genome sequence-based species classification scheme for Morganella is established with a higher resolution and accuracy than those of the conventional scheme. Second, the prevalence of accessory genetic elements (AGEs) and associated antimicrobial resistance genes (ARGs) among Morganella isolates is disclosed based on genome sequences. Finally, a detailed sequence comparison of eight groups of 23 AGEs (including 19 Morganella chromosomal AGEs) reveals that Morganella chromosomes have evolved to acquire diverse AGEs harboring different profiles of ARGs and that some of these AGEs harbor large accessory modules that exhibit complex mosaic structures and contain a large number of ARGs. Data presented here provide a deeper understanding of the classification and evolution of Morganella species and also those of ARG-containing AGEs in Morganella at the genomic scale.

ProtPlat: an efficient pre-training platform for protein classification based on FastText

BackgroundFor the past decades, benefitting from the rapid growth of protein sequence data in public databases, a lot of machine learning methods have been developed to predict physicochemical properties or functions of proteins using amino acid sequence features. However, the prediction performance often suffers from the lack of labeled data. In recent years, pre-training methods have been widely studied to address the small-sample issue in computer vision and natural language processing fields, while specific pre-training techniques for protein sequences are few.ResultsIn this paper, we propose a pre-training platform for representing protein sequences, called ProtPlat, which uses the Pfam database to train a three-layer neural network, and then uses specific training data from downstream tasks to fine-tune the model. ProtPlat can learn good representations for amino acids, and at the same time achieve efficient classification. We conduct experiments on three protein classification tasks, including the identification of type III secreted effectors, the prediction of subcellular localization, and the recognition of signal peptides. The experimental results show that the pre-training can enhance model performance effectively and ProtPlat is competitive to the state-of-the-art predictors, especially for small datasets. We implement the ProtPlat platform as a web service (https://compbio.sjtu.edu.cn/protplat) that is accessible to the public.ConclusionsTo enhance the feature representation of protein amino acid sequences and improve the performance of sequence-based classification tasks, we develop ProtPlat, a general platform for the pre-training of protein sequences, which is featured by a large-scale supervised training based on Pfam database and an efficient learning model, FastText. The experimental results of three downstream classification tasks demonstrate the efficacy of ProtPlat.

Implementation of Sequence-Based Classification Methods for Motion Assessment and Recognition in a Traditional Chinese Sport (Baduanjin).

This study aimed to assess the motion accuracy of Baduanjin and recognise the motions of Baduanjin based on sequence-based methods. Motion data of Baduanjin were measured by the inertial sensor measurement system (IMU). Fifty-four participants were recruited to capture motion data. Based on the motion data, various sequence-based methods, namely dynamic time warping (DTW) combined with classifiers, hidden Markov model (HMM), and recurrent neural networks (RNNs), were applied to assess motion accuracy and recognise the motions of Baduanjin. To assess motion accuracy, the scores for motion accuracies from teachers were used as the standard to train the models on the different sequence-based methods. The effectiveness of Baduanjin motion recognition with different sequence-based methods was verified. Among the methods, DTW + k-NN had the highest average accuracy (83.03%) and shortest average processing time (3.810 s) during assessing. In terms of motion reorganisation, three methods (DTW + k-NN, DTW + SVM, and HMM) had the highest accuracies (over 99%), which were not significantly different from each other. However, the processing time of DTW + k-NN was the shortest (3.823 s) compared to the other two methods. The results show that the motions of Baduanjin could be recognised, and the accuracy can be assessed through an appropriate sequence-based method with the motion data captured by IMU.

How many α-amylase GH families are there in the CAZy database?

Abstract The CAZy database is a web-server for sequence-based classification of carbohydrate-active enzymes that has become the worldwide and indispensable tool for scientists engaged in this research field. It was originally created in 1991 as a classification of glycoside hydrolases (GH) and currently, this section of CAZy represents its largest part counting 172 GH families. The present Opinion paper is devoted to the specificity of α-amylase (EC 3.2.1.1) and its occurrence in the CAZy database. Among the 172 defined GH families, four, i.e. GH13, GH57, GH119 and GH126, may be considered as the α-amylase GH families. This view reflects a historical background and traditions widely accepted during the previous decades with respect to the chronology of creating the individual GH families. It obeys the phenomenon that some amylolytic enzymes, which were used to create the individual GH families and were originally known as α-amylases, according to current knowledge from later, more detailed characterization, need not necessarily represent genuine α-amylases. Our Opinion paper was therefore written in an effort to invite the scientific community to think about that with a mind open to changes and to consider the seemingly unambiguous question in the title as one that may not have a simple answer.

NoAS-DS: Neural optimal architecture search for detection of diverse DNA signals

Neural network architectures are high-performing variable models that can solve many learning tasks. Designing architectures manually require substantial time and also prior knowledge and expertise to develop a high-accuracy model. Most of the architecture search methods are developed over the task of image classification resulting in the building of complex architectures intended for large data inputs such as images. Motivated by the applications of DNA computing in Neural Architecture Search (NAS), we propose NoAS-DS which is specifically built for the architecture search of sequence-based classification tasks. Furthermore, NoAS-DS is applied to the task of predicting binding sites. Unlike other methods that implement only Convolution layers, NoAS-DS, specifically combines Convolution and LSTM layers that helps in the process of automatic architecture building. This hybrid approach helped in achieving high accuracy results on TFBS and RBP datasets which outperformed other models in TF-DNA binding prediction tasks. The best architectures generated by the proposed model can be applied to other DNA datasets of similar nature using transfer learning technique that demonstrates its generalization capability. This greatly reduces the effort required to build new architectures for other prediction tasks.

New Insight regarding Legionella Non-Pneumophila Species Identification: Comparison between the Traditional mip Gene Classification Scheme and a Newly Proposed Scheme Targeting the rpoB Gene.

ABSTRACTThe identification of Legionella non-pneumophila species (non-Lp) in clinical and environmental samples is based on the mip gene, although several studies suggest its limitations and the need to expand the classification scheme to include other genes. In this study, the development of a new classification scheme targeting the rpoB gene is proposed to obtain a more reliable identification of 135 Legionella environmental isolates. All isolates were sequenced for the mip and rpoB genes, and the results were compared to study the discriminatory power of the proposed rpoB scheme. Complete concordance between the mip and rpoB results based on genomic percent identity was found for 121/135 (89.6%) isolates; in contrast, discordance was found for 14/135 (10.4%) isolates. Additionally, due to the lack of reference values for the rpoB gene, inter- and intraspecies variation intervals were calculated based on a pairwise identity matrix that was built using the entire rpoB gene (∼4,107 bp) and a partial region (329 bp) to better evaluate the genomic identity obtained. The interspecies variation interval found here (4.9% to 26.7%) was then proposed as a useful sequence-based classification scheme for the identification of unknown non-Lp isolates. The results suggest that using both the mip and rpoB genes makes it possible to correctly discriminate between several species, allowing possible new species to be identified, as confirmed by preliminary whole-genome sequencing analyses performed on our isolates. Therefore, starting from a valid and reliable identification approach, the simultaneous use of mip and rpoB associated with other genes, as it occurs with the sequence-based typing (SBT) scheme developed for Legionella pneumophila, could support the development of multilocus sequence typing to improve the knowledge and discovery of Legionella species subtypes.IMPORTANCE Legionella spp. are a widely spread bacteria that cause a fatal form of pneumonia. While traditional laboratory techniques have provided valuable systems for Legionella pneumophila identification, the amplification of the mip gene has been recognized as the only useful tool for Legionella non-pneumophila species identification both in clinical and environmental samples. Several studies focused on the mip gene classification scheme showed its limitations and the need to improve the classification scheme, including other genes. Our study provides significant advantages on Legionella identification, providing a reproducible new rpoB gene classification scheme that seems to be more accurate than mip gene sequencing, bringing out greater genetic variation on Legionella species. In addition, the combined use of both the mip and rpoB genes allowed us to identify presumed new Legionella species, improving epidemiological investigations and acquiring new understanding on Legionella fields.