Microbiome Composition and Function in Aquatic Vertebrates: Small Organisms Making Big Impacts on Aquatic Animal Health

Frontiers in Microbiology, Journal Year: 2021, Volume and Issue: 12

Published: March 11, 2021

Aquatic ecosystems are under increasing stress from global anthropogenic and natural changes, including climate change, eutrophication, ocean acidification, pollution. In this critical review, we synthesize research on the microbiota of aquatic vertebrates discuss impact emerging stressors microbial communities using two case studies, that toxic cyanobacteria microplastics. Most studies to date focused host-associated microbiomes individual organisms, however, few take an integrative approach examine vertebrate by considering both free-living within ecosystem. We highlight what is known about in ecosystems, with a focus interface between water, fish, marine mammals. Though water vary geography, temperature, depth, other factors, core functions such as primary production, nitrogen cycling, nutrient metabolism often conserved across environments. outline knowledge composition function tissue-specific fish mammals environmental factors influencing their structure. The highly unique species delicate balance respiratory, skin, gastrointestinal exists host. vertebrates, conditions ecological niche driving behind function. also generate comprehensive catalog mammal genera, revealing commonalities among species, potential use indicators health status ecosystems. importance functional relevance relation organism physiology ability overcome related change. Understanding dynamic relationship animals they colonize for monitoring quality population health.

Language: Английский

---

gutMGene: a comprehensive database for target genes of gut microbes and microbial metabolites

Nucleic Acids Research, Journal Year: 2021, Volume and Issue: 50(D1), P. D795 - D800

Published: Sept. 8, 2021

Abstract gutMGene (http://bio-annotation.cn/gutmgene), a manually curated database, aims at providing comprehensive resource of target genes gut microbes and microbial metabolites in humans mice. Metagenomic sequencing fecal samples has identified 3.3 × 106 non-redundant from up to 1500 different species. One the contributions microbiota host biology is circulating pool bacterially derived small-molecule metabolites. It been estimated that 10% found mammalian blood are microbiota, where they can produce systemic effects on through activating or inhibiting gene expression. The current version documents 1331 relationships between 332 microbes, 207 223 humans, 2349 209 149 544 Each entry contains detailed information relationship microbe, metabolite gene, brief description relationship, experiment technology platform, literature reference so on. provides user-friendly interface browse retrieve each using disorders intervention measures. also offers option download all entries submit new experimentally validated associations.

Language: Английский

---

Anticancer peptides prediction with deep representation learning features

Briefings in Bioinformatics, Journal Year: 2021, Volume and Issue: 22(5)

Published: Jan. 6, 2021

Abstract Anticancer peptides constitute one of the most promising therapeutic agents for combating common human cancers. Using wet experiments to verify whether a peptide displays anticancer characteristics is time-consuming and costly. Hence, in this study, we proposed computational method named identify via deep representation learning features (iACP-DRLF) using light gradient boosting machine algorithm features. Two kinds sequence embedding technologies were used, namely soft symmetric alignment unified (UniRep) embedding, both which involved neural network models based on long short-term memory networks their derived networks. The results showed that use greatly improved capability discriminate from other peptides. Also, UMAP (uniform manifold approximation projection dimension reduction) SHAP (shapley additive explanations) analysis proved UniRep have an advantage over identification. python script pretrained could be downloaded https://github.com/zhibinlv/iACP-DRLF or http://public.aibiochem.net/iACP-DRLF/.

Language: Английский

---

sgRNACNN: identifying sgRNA on-target activity in four crops using ensembles of convolutional neural networks

Plant Molecular Biology, Journal Year: 2021, Volume and Issue: 105(4-5), P. 483 - 495

Published: Jan. 1, 2021

Language: Английский

---

A First Computational Frame for Recognizing Heparin-Binding Protein

Diagnostics, Journal Year: 2023, Volume and Issue: 13(14), P. 2465 - 2465

Published: July 24, 2023

Heparin-binding protein (HBP) is a cationic antibacterial derived from multinuclear neutrophils and an important biomarker of infectious diseases. The correct identification HBP great significance to the study This work provides first recognition framework based on machine learning accurately identify HBP. By using four sequence descriptors, non-HBP samples were represented by discrete numbers. inputting these features into support vector (SVM) random forest (RF) algorithm comparing prediction performances methods training data independent test data, it found that SVM-based classifier has greatest potential model could produce auROC 0.981 ± 0.028 10-fold cross-validation overall accuracy 95.0% data. As for recognition, will provide some help diseases stimulate further research in related fields.

Language: Английский

---

GMrepo v2: a curated human gut microbiome database with special focus on disease markers and cross-dataset comparison

Nucleic Acids Research, Journal Year: 2021, Volume and Issue: 50(D1), P. D777 - D784

Published: Oct. 13, 2021

GMrepo (data repository for Gut Microbiota) is a database of curated and consistently annotated human gut metagenomes. Its main purposes are to increase the reusability accessibility metagenomic data, enable cross-project phenotype comparisons. To achieve these goals, we performed manual curation on meta-data organized datasets in phenotype-centric manner. v2 contains 353 projects 71,642 runs/samples, which significantly increased from previous version. Among 45,111 26,531 were obtained by 16S rRNA amplicon whole-genome metagenomics sequencing, respectively. We also number phenotypes 92 133. In addition, introduced disease-marker identification cross-project/phenotype comparison. first identified disease markers between two (e.g. health versus diseases) per-project basis selected projects. then compared each pair across facilitate consistent microbial datasets. Finally, provided marker-centric view allow users check if marker has different trends diseases. So far, includes 592 taxa (350 species 242 genera) 47 pairs, 83 freely available at: https://gmrepo.humangut.info.

Language: Английский

---

GANsDTA: Predicting Drug-Target Binding Affinity Using GANs

Frontiers in Genetics, Journal Year: 2020, Volume and Issue: 10

Published: Jan. 9, 2020

The computational prediction of interactions between drugs and targets is a standing challenge in drug discovery. State-of-the-art methods for drug-target interaction are primarily based on supervised machine learning with known labels information. However, biomedicine, obtaining labeled training data an expensive laborious process. This paper proposes semi-supervised generative adversarial networks (GANs)-based method to predict binding affinity. Our comprises two parts, GANs feature extraction regression network prediction. mechanism allows our model learn proteins features both unlabled data. We evaluate the performance using multiple public datasets. Experimental results demonstrate that achieves competitive while utilizing freely available unlabeled suggest such can considerably help improve various biomedical relation processes, example, Drug-Target protein-protein interaction, particularly when only limited tasks.To best knowledge, this first GANs-based

Language: Английский

---

Dysbiosis of Gut Microbiota in Patients With Acute Myocardial Infarction

Frontiers in Microbiology, Journal Year: 2021, Volume and Issue: 12

Published: July 5, 2021

Acute myocardial infarction (AMI) continues as the main cause of morbidity and mortality worldwide. Interestingly, emerging evidence highlights role gut microbiota in regulating pathogenesis coronary heart disease, but few studies have systematically assessed alterations influence AMI patients. As one approach to address this deficiency, study composition fecal microflora was determined from Chinese patients links between clinical features functional pathways were assessed. Fecal samples 30 healthy controls collected identify using bacterial 16S rRNA gene sequencing. We found that contained a lower abundance phylum

Language: Английский

---

RF-PseU: A Random Forest Predictor for RNA Pseudouridine Sites

Frontiers in Bioengineering and Biotechnology, Journal Year: 2020, Volume and Issue: 8

Published: Feb. 25, 2020

One of the ubiquitous chemical modifications in RNA, pseudouridine modification is crucial for various cellular biological and physiological processes. To gain more insight into functional mechanisms involved, it fundamental importance to precisely identify sites RNA. Several useful machine learning approaches have become available recently, with increasing progress next-generation sequencing technology; however, existing methods cannot predict high accuracy. Thus, a accurate predictor required. In this study, random forest-based named RF-PseU proposed prediction pseudouridylation sites. optimize feature representation obtain better model, light gradient boosting algorithm incremental selection strategy were used select optimum space vector training forest model RF-PseU. Compared previous state-of-the-art predictors, results on same benchmark data sets three species demonstrate that performs overall. The integrated average leave-one-out cross-validation independent testing accuracy scores 71.4% 74.7%, respectively, representing increments 3.63% 4.77% versus best predictor. Moreover, final was built provides reliable robust tool identifying A web server user-friendly interface accessible at http://148.70.81.170:10228/rfpseu.

Language: Английский

---

Human reference gut microbiome catalog including newly assembled genomes from under-represented Asian metagenomes

Genome Medicine, Journal Year: 2021, Volume and Issue: 13(1)

Published: Aug. 26, 2021

Abstract Background Metagenome sampling bias for geographical location and lifestyle is partially responsible the incomplete catalog of reference genomes gut microbial species. Thus, genome assembly from currently under-represented populations may effectively expand microbiome improve taxonomic functional profiling. Methods We assembled using public whole-metagenomic shotgun sequencing (WMS) data 110 645 fecal samples India Japan, respectively. In addition, we newly generated WMS 90 collected Korea. Expecting low-abundance species require a much deeper than that usually employed, so performed ultra-deep (> 30 Gbp or > 100 million read pairs) consequently 29,082 prokaryotic 845 metagenomes three Asian countries combined them with Unified Human Gastrointestinal Genome (UHGG) to generate an expanded catalog, Reference Gut Microbiome (HRGM). Results HRGM contains 232,098 non-redundant 5414 representative including 780 are novel, 103 unique proteins, 274 single-nucleotide variants. This over 10% increase UHGG. The new were enriched Bacteroidaceae family, associated high-fiber seaweed-rich diets. Single-nucleotide variant density was positively speciation rate commensals. found facilitated taxa, deep (e.g., 20 be needed profiling taxa. Importantly, significantly improved classification reads samples. Finally, analysis human self-antigen homologs on suggested bacterial taxa high cross-reactivity potential contribute more pathogenesis microbiome-associated diseases those low by promoting inflammatory condition. Conclusions By previously countries, Korea, India, developed substantially HRGM. Information coding genes publicly available ( www.mbiomenet.org/HRGM/ ). will facilitate identification disease-associated microbiota.

Language: Английский

---