Biology subjects

Coelho, L. P.

Publications and source records attributed to Coelho, L. P..

18 recordsLinked to original sources

bioRxiv2026

The elusive resistome: a global comparison reveals large discrepancies among detection pipelines

Identifying antibiotic resistance genes (ARGs) from metagenomic data is critical for studying antimicrobial resistance across microbial communities and pathogens. However, there is no standardized methodology for ARG annotation. Here, we compare ten commonly used ARG detection pipelines by analysing over 270 million prokaryotic genes from the Global Microbial Gene Catalogue across 13 distinct habitats. We observed up to a 45-fold difference in the number of reported ARGs, with a mean Jaccard index of only 16% between pipelines. Pipeline selection profoundly impacted downstream biological interpretations, with drastic changes to estimates of ARG relative abundance and richness, to the characterization of pan- and core-resistomes, and to the class-level composition of the inferred resistome. ARG detection pipelines make different, defensible trade-offs, and no single approach should be treated as authoritative. Therefore, users should justify and communicate choices carefully, as our analyses show that, taken uncritically, the same data can support conflicting biological and ecological interpretations.

bioinformatics↗

bioRxiv2026

Long-read metagenomic sequencing reveals novel lineages and functional diversity in urban soil microbiome

City parks and other urban green spaces can bring significant benefits to the physical and mental health of city residents. However, there is limited knowledge about the microbial communities inhabiting these urban soils. Here, we applied long-read metagenomic sequencing to 58 urban soil samples from two major cities in China, enabling genome-resolved reconstruction of microbial diversity at unprecedented contiguity. We recovered 7,949 medium- and high-quality metagenome-assembled genomes, comprising 4,171 species-level genome bins, of which over 97% represent previously undescribed species. Long-read assemblies revealed extensive secondary metabolic capacity, including more than 30,000 biosynthetic gene clusters, which were highly contiguous compared with those from fragmented short-read assemblies. Beyond secondary metabolism, we uncovered over 2 million small protein families, including hundreds that are strongly enriched in the neighbourhood of defense systems and mobile genetic elements, highlighting their overlooked role in urban soils. These findings expand our understanding of the functional diversity of urban soil microbiomes and provide new insights with implications for urban public health.

bioinformatics↗

bioRxiv2025

Capturing global pet dog gut microbial diversity and hundreds of near-finished bacterial genomes by using long-read metagenomics in a Shanghai cohort

Pet dogs are considered part of the family, and understanding their gut microbiomes can provide insights into both animal and household health. Most comprehensive studies, however, relied on short-read sequencing, resulting in fragmented MAGs that miss mobile elements, antimicrobial-resistance genes, and ribosomal genes. Here, we applied deep long-read metagenomics (polished with short-reads) to fecal samples from 51 urban pet dogs in Shanghai, generating 2,676 MAGs--representing 320 bacterial species--, of which [~]72% achieved near-finished quality, often improving on the corresponding reference public genome. Comparisons with external datasets showed that our Shanghai-based MAG catalog is representative of pet dogs worldwide (median read mapping of >90%). Moreover, we recovered circular extrachromosomal elements, including those linked to antimicrobial resistance, which were also detected in external dog gut datasets. In conclusion, we provide a high-quality reference resource and demonstrate the power of deep long-read metagenomics to resolve microbial diversity in complex host-associated microbiomes.

microbiology↗

bioRxiv2025

AEMB: a computationally efficient abundance estimation method for metagenomic binning

Metagenomic binning is a crucial step in metagenomic analysis, namely grouping together contigs that are predicted to originate from the same genome to enable the recovery of metagenome-assembled genomes (MAGs). It has been shown that using information from multiple samples yields better results than binning each sample independently. However, for N metagenomic samples, using full multi-against-multi binning requires N2 alignments, making it computationally challenging to apply in large-scale metagenomic studies. Here, we propose AEMB (Abundance Estimation for Metagenomic Binning), a novel mapping mode implemented in strobealign. AEMB is a computationally efficient abundance estimation method that uses a prefix-lookup vector as an indexing structure to reduce memory usage and randstrobes to estimate the abundance of contigs without performing base-level alignment. Compared to the hash table used in the previous version of strobealign, the indexing structure reduces peak memory usage by 25.2% with almost the same runtime. Furthermore, we implemented a fast abundance estimation method that skips base-level alignment. Altogether, AEMB reduces the runtime for abundance estimation by 88% to 96% compared to commonly used alignment methods such as Bowtie2 and BWA, while achieving similar binning results. AEMB is available as a mapping mode in strobealign https://github.com/ksahlin/strobealign and SemiBin2 (v2.1 and later) accepts its inputs for binning.

bioinformatics↗

bioRxiv2025

A census of hidden and discoverable microbial diversity beyond genome-centric approaches

Cataloguing Earths biodiversity remains one of the most formidable challenges in biology, and the greatest diversity is expected to reside among the smallest organisms: microbes. Yet the ongoing census of microbial life is hampered by disparate sampling of Earths habitats, challenges in isolating uncultivated organisms, limited resolution in taxonomic marker gene amplicons, and incomplete recovery of metagenome-assembled genomes (MAGs). Here, we quantified discoverable bacterial and archaeal diversity in a comprehensive, curated cross-habitat dataset of 92,187 metagenomes. Clustering 502M sequences of 130 marker genes, we detected 705k bacterial and 27k archaeal species-level clades, the vast majority of which was hidden among unbinned contigs. At deeper taxonomic levels, we estimate that 10 archaeal and 145 bacterial novel phyla and around 80k novel genera are discoverable in current data. We identified soils and aquatic environments as novel lineage recovery hotspots, yet predict that discovery will remain in full swing across habitats as more data accrues. Finally, we show that prokaryotic diversity follows power laws, confirming century-old hypotheses on clade size patterns and suggesting that novel lineages arise within common (and fractal) evolutionary patterns, comparable to those among eukaryotic clades and viruses, along the full depth and breadth of the Tree of Life.

microbiology↗

bioRxiv2024

Ubiquity of inverted 'gelatinous' ecosystem pyramids in the global ocean

Summary paragraphPlankton are essential in marine ecosystems. However, our knowledge of overall community structure is sparse due to inconsistent sampling across their very large organismal size range. Here we use diverse imaging methods to establish complete plankton inventories of organisms spanning five orders of magnitude in size. Plankton community size and trophic structure variation validate a long-held theoretical link between organism size-spectra and ecosystem trophic structures. We found that predator/grazer biomass and biovolume unexpectedly exceed that of primary producers at most (55%) locations, likely due to our better quantification of gelatinous organisms. Bottom- heavy ecosystems (the norm on land) appear to be rare in the ocean. Collectively, gelatinous organisms represent 30% of the total biovolume (8-9% of carbon) of marine plankton communities from tropical to polar ecosystems. Communities can be split into three extreme typologies: diatom/copepod-dominated in eutrophic blooms, rhizarian/chaetognath-dominated in oligotrophic tropical oceans, and gelatinous-dominated elsewhere. While plankton taxonomic composition changes with latitude, functional and trophic structures mostly depend on the amount of prey available for each trophic level. Given future projections of oligotrophication of marine ecosystems, our findings suggest that rhizarian and gelatinous organisms will increasingly dominate the apex position of planktonic ecosystems, leading to significant changes in the oceans carbon cycle.

ecology↗

bioRxiv2023

A catalogue of small proteins from the global microbiome

Small open reading frames (smORFs) shorter than 100 codons are widespread and perform essential roles in microorganisms, where they encode proteins active in several cell functions, including signal pathways, stress response, and antibacterial activities. However, the ecology, distribution and role of small proteins in the global microbiome remain unknown. Here, we constructed a global microbial smORFs catalogue (GMSC) derived from 63,410 publicly available metagenomes across 75 distinct habitats and 87,920 high-quality isolate genomes. GMSC contains 965 million non-redundant smORFs with comprehensive annotations. We found that archaea harbor more small proteins proportionally than bacteria. We moreover provide a tool called GMSC-mapper to identify and annotate small proteins from microbial (meta)genomes. Overall, this publicly-available resource demonstrates the immense and underexplored diversity of small proteins.

bioinformatics↗

bioRxiv2023

SemiBin2: self-supervised contrastive learning leads to better MAGs for short- and long-read sequencing

MotivationMetagenomic binning methods to reconstruct metagenome-assembled genomes (MAGs) from environmental samples have been widely used in large-scale metagenomic studies. The recently proposed semi-supervised binning method, SemiBin, achieved state-of-the-art binning results in several environments. However, this required annotating contigs, a computationally costly and potentially biased process. ResultsWe propose SemiBin2, which uses self-supervised learning to learn feature embeddings from the contigs. In simulated and real datasets, we show that self-supervised learning achieves better results than the semi-supervised learning used in SemiBin1 and that SemiBin2 outperforms other state-of-the-art binners. Compared to SemiBin1, SemiBin2 can reconstruct 8.3%-21.5% more high-quality bins and requires only 25% of the running time and 11% of peak memory usage in real short-read sequencing samples. To extend SemiBin2 to long-read data, we also propose ensemble-based DBSCAN clustering algorithm, resulting in 13.1%-26.3% more high-quality genomes than the second best binner for long-read data. Availability and ImplementationSemiBin2 is available as open source software at https://github.com/BigDataBiology/SemiBin/ and the analysis scripts used in the study can be found at https://github.com/BigDataBiology/SemiBin2_benchmark.

bioinformatics↗

bioRxiv2022

ResFinderFG v2.0: a database of antibiotic resistance genes obtained by functional metagenomics

Metagenomics can be used to monitor the spread of antibiotic resistance genes (ARGs). ARGs found in databases such as ResFinder and CARD primarily originate from culturable and pathogenic bacteria, while ARGs from non-culturable and non-pathogenic bacteria remain understudied. Functional metagenomics is based on phenotypic gene selection and can identify ARGs from non-culturable bacteria with a potentially low identity shared with known ARGs. In 2016, the ResFinderFG v1.0 database was created to collect ARGs from functional metagenomics studies. Here, we present the second version of the database, ResFinderFG v2.0, which is available on the Center of Genomic Epidemiology web server (https://cge.food.dtu.dk/services/ResFinderFG/). It comprises 3,913 ARGs identified by functional metagenomics from 50 carefully curated datasets. We assessed its potential to detect ARGs in comparison to other popular databases in gut, soil and water (marine + freshwater) Global Microbial Gene Catalogues (https://gmgc.embl.de). ResFinderFG v2.0 allowed for the detection of ARGs that were not detected using other databases. These included ARGs conferring resistance to beta-lactams, cycline, phenicol, glycopeptide/cycloserine and trimethoprim/sulfonamide. Thus, ResFinderFG v2.0 can be used to identify ARGs differing from those found in conventional databases and therefore improve the description of resistomes. GRAPHICAL ABSTRACT O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=137 SRC="FIGDIR/small/512667v2_ufig1.gif" ALT="Figure 1"> View larger version (36K): org.highwire.dtl.DTLVardef@1a16f6dorg.highwire.dtl.DTLVardef@19a7589org.highwire.dtl.DTLVardef@19a7d5forg.highwire.dtl.DTLVardef@19d1c16_HPS_FORMAT_FIGEXP M_FIG C_FIG Caption: Additional use of ResFinderFG v2.0 database (composed of antibiotic resistance genes obtained with functional metagenomics) on the Center of Genomic Epidemiology webserver (https://cge.food.dtu.dk/services/ResFinderFG/), allows for more exhaustive resistome descriptions.

microbiology↗

bioRxiv2022

Functional and evolutionary significance of unknown genes from uncultivated taxa

Most microbes on our planet remain uncultured and poorly studied. Recent efforts to catalog their genetic diversity have revealed that a significant fraction of the observed microbial genes are functional and evolutionary untraceable, lacking homologs in reference databases. Despite their potential biological value, these apparently unrelated orphan genes from uncultivated taxa have been routinely discarded in metagenomics surveys. Here, we analyzed a global multi-habitat dataset covering 151,697 medium and high-quality metagenome assembled genomes (MAGs), 5,969 single-amplified genomes (SAGs), and 19,642 reference genomes, and identified 413,335 highly curated novel protein families under strong purifying selection out of previously considered orphan genes. These new protein families, representing a three-fold increase over the total number of prokaryotic orthologous groups described to date, spread out across the prokaryote phylogeny, can span multiple habitats, and are notably overrepresented in recently discovered taxa. By genomic context analysis, we pinpointed thousands of unknown protein families to phylogenetically conserved operons linked to energy production, xenobiotic metabolism and microbial resistance. Most remarkably, we found 980 previously neglected protein families that can accurately distinguish entire uncultivated phyla, classes, and orders, likely representing synapomorphic traits that fostered their divergence. The systematic curation and evolutionary analysis of the unique genetic repertoire of uncultivated taxa opens new avenues for understanding the biology and ecological roles of poorly explored lineages at a global scale.

microbiology↗

bioRxiv2021

Drivers and Determinants of Strain Dynamics Following Faecal Microbiota Transplantation

Faecal microbiota transplantation (FMT) is an efficacious therapeutic intervention, but its clinical mode of action and underlying microbiome dynamics remain poorly understood. Here, we analysed the metagenomes associated with 142 FMTs, in a time series-based meta-study across five disease indications. We quantified strain-level dynamics of 1,089 microbial species based on their pangenome, complemented with 47,548 newly constructed metagenome-assembled genomes. Using subsets of procedural-, host- and microbiome-based variables, LASSO-regularised regression models accurately predicted the colonisation and resilience of donor and recipient microbes, as well as turnover of individual species. Linking this to putative ecological mechanisms, we found these sets of variables to be informative of the underlying processes that shape the post-FMT gut microbiome. Recipient factors and complementarity of donor and recipient microbiomes, encompassing entire communities to individual strains, were the main determinants of individual strain population dynamics, and mostly independent of clinical outcomes. Recipient community state and the degree of residual strain depletion provided a neutral baseline for donor strain colonisation success, in addition to inhibitive priority effects between species and conspecific strains, as well as putatively adaptive processes. Our results suggest promising tunable parameters to enhance donor flora colonisation or recipient flora displacement in clinical practice, towards the development of more targeted and personalised therapies.

microbiology↗

bioRxiv2021

SemiBin: Incorporating information from reference genomes with semi-supervised deep learning leads to better metagenomic assembled genomes (MAGs)

Metagenomic binning is the step in building metagenome-assembled genomes (MAGs) when sequences predicted to originate from the same genome are automatically grouped together. The most widely-used methods for binning are reference-independent, operating de novo and allow the recovery of genomes from previously unsampled clades. However, they do not leverage the knowledge in existing databases. Here, we propose SemiBin, an open source tool that uses neural networks to implement a semi-supervised approach, i.e. SemiBin exploits the information in reference genomes, while retaining the capability of binning genomes that are outside the reference dataset. SemiBin outperforms existing state-of-the-art binning methods in simulated and real microbiome datasets across three different environments (human gut, dog gut, and marine microbiomes). SemiBin returns more high-quality bins with larger taxonomic diversity, including more distinct genera and species. SemiBin is available as open source software at https://github.com/BigDataBiology/SemiBin/.

bioinformatics↗

bioRxiv2021

metaMIC: reference-free Misassembly Identification and Correction of de novo metagenomic assemblies

Evaluating the quality of metagenomic assemblies is important for constructing reliable metagenome-assembled genomes and downstream analyses. Here, we present metaMIC (https://github.com/ZhaoXM-Lab/metaMIC), a machine-learning based tool for identifying and correcting misassemblies in metagenomic assemblies. Benchmarking results on both simulated and real datasets demonstrate that metaMIC outperforms existing tools when identifying misassembled contigs. Furthermore, metaMIC is able to localize the misassembly breakpoints, and the correction of misassemblies by splitting at misassembly breakpoints can improve downstream scaffolding and binning results.

bioinformatics↗

bioRxiv2020

GUNC: Detection of Chimerism and Contamination in Prokaryotic Genomes

Genomes are critical units in microbiology, yet ascertaining quality in prokaryotic genomes remains a formidable challenge. We present GUNC (the Genome UNClutterer), a tool that accurately detects and quantifies genome chimerism based on the lineage homogeneity of individual contigs using a genomes full complement of genes. GUNC complements existing approaches by targeting previously underdetected types of contamination: we conservatively estimate that 5.7% of genomes in GenBank, 5.2% in RefSeq, and 15-30% of pre-filtered high quality metagenome-assembled genomes in recent studies are undetected chimeras. GUNC provides a fast and robust tool to substantially improve prokaryotic genome quality. Source code (GPLv3+): https://github.com/grp-bork/gunc

bioinformatics↗

bioRxiv2019

FACS: antimicrobial peptide screening in genomes and metagenomes

MotivationAntimicrobial peptides (AMPs) have the potential to tackle multidrug-resistant pathogens in both clinical and non-clinical contexts. The recent growth in the availability of genomes and metagenomes provides an opportunity for in silico prediction of novel AMP molecules. However, due to the small size of these peptides, standard gene prospection methods cannot be applied in this domain and alternative approaches are necessary. In particular, standard gene prediction methods have low precision for short peptides, and functional classification by homology results in low recall. ResultsHere, we present Macrel (for metagenomic AMP classification and retrieval), which is an end-to-end pipeline for the prospection of high-quality AMP candidates from (meta)genomes. For this, we introduce a novel set of 22 peptide features. These were used to build classifiers which perform similarly to the state-of-the-art in the prediction of both antimicrobial and hemolytic activity of peptides, but with enhanced precision (using standard benchmarks as well as a stricter testing regime). We demonstrate that Macrel recovers high-quality AMP candidates using realistic simulations and real data. AvailabilityMacrel is implemented in Python 3. It is available as open source at https://github.com/BigDataBiology/macrel and through bioconda. Classification of peptides or prediction of AMPs in contigs can also be performed on the webserver: http://big-data-biology.org/software/macrel.

bioinformatics↗

bioRxiv2019

Disentangling the impact of environmental and phylogenetic constraints on prokaryotic strain diversity

Microbial organisms inhabit virtually all environments and encompass a vast biological diversity. The pan-genome concept aims to facilitate an understanding of diversity within defined phylogenetic groups. Hence, pan-genomes are increasingly used to characterize the strain diversity of prokaryotic species. To understand the interdependency of pan-genome features (such as numbers of core and accessory genes) and to study the impact of environmental and phylogenetic constraints on the evolution of conspecific strains, we computed pan-genomes for 155 phylogenetically diverse species using 7000 high-quality genomes. We show that many pan-genome features such as functional diversity and core genome nucleotide diversity are correlated to each other. Further, habitat flexibility as approximated by species ubiquity is associated with several pan-genome features, particularly core genome size. In general, environment had a stronger impact on pan-genome features than phylogenetic signal. Similar environmental preferences led to convergent evolution of pan-genomic features in distant phylogenetic clades. For example, the soil environment promotes expansion of pan-genome size, while host-associated habitats lead to its reduction.

genomics↗

bioRxiv2018

Extensive Transmission of Microbes along the Gastrointestinal Tract

The gastrointestinal tract is abundantly colonized by microbes, yet the translocation of oral species to the intestine is considered a rare aberrant event, and a hallmark of disease. By studying salivary and fecal microbial strain populations of 310 species in 470 individuals from five countries, we found that transmission to, and subsequent colonization of, the large intestine by oral microbes is common and extensive among healthy individuals. We found evidence for a vast majority of oral species to be transferable, with increased levels of transmission in colorectal cancer and rheumatoid arthritis patients and, more generally, for species described as opportunistic pathogens. This establishes the oral cavity as an endogenous reservoir for gut microbial strains, and oral-fecal transmission as an important process that shapes the gastrointestinal microbiome in health and disease.

microbiology↗

bioRxiv2018

NG-meta-profiler: fast processing of metagenomes using NGLess, a domain-specific language

NGLess is a domain specific language for describing next-generation sequence processing pipelines. It was developed with the goal of enabling user-friendly computational reproducibility.\n\nUsing this framework, we developed NG-meta-profiler, a fast profiler for metagenomes which performs sequence preprocessing, mapping to bundled databases, filtering of the mapping results, and profiling (taxonomic and functional). It is significantly faster than either MOCAT2 or htseq-count and (as it builds on NGLess) its results are perfectly reproducible. These pipelines can easily be customized and extended with other tools.\n\nNGLess and NG-meta-profiler are open source software (under the liberal MIT licence) and can be downloaded from http://ngless.embl.de or installed through bioconda.

bioinformatics↗