Pacbio HiFi sequencing sheds light on key bacteria contributing to deadwood decomposition processes

Pacbio HiFi sequencing sheds light on key bacteria contributing to deadwood decomposition processes | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Pacbio HiFi sequencing sheds light on key bacteria contributing to deadwood decomposition processes Etienne Richy, Priscila Thiago Dobbler, Vojtěch Tláskal, Rubén López-Mondéjar, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4181686/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background In forest ecosystems, biological decomposition of deadwood components plays a pivotal role in nutrient cycling and in carbon storage by enriching soils with organic matter. However, deciphering the functional features of deadwood microbiomes is challenging due to their complexity and the limitations of traditional cultivation methods. Our study demonstrates how such limitations can be overcome by describing metagenome composition and function through the analysis of long DNA molecules using the PacBio HiFi platform. Results The accuracy of PacBio HiFi long-read sequencing emerges as a robust tool for reconstructing microbial genomes in deadwood. It outperformed the routine short-read sequencing and genome sequencing of isolates in terms of the numbers of genomes recovered, their completeness, and representation of their functional potential. We successfully assembled 69 bacterial genomes representing seven out of eight predominant bacterial phyla, including 14 high-quality draft MAGs and 7 nearly finished MAGs. Notably, the genomic exploration extends to Myxococcota, unveiling the unique capacity of Polyangiaceae to degrade cellulose. Patescibacteria contributed to deadwood decomposition processes, actively decomposing hemicellulose and recycling fungal-derived compounds. Furthermore, a novel nitrogen-fixing bacteria within the Steroidobacteriaceae family were identified, displaying interesting genomic adaptations to environmental conditions. The discovered diversity of biosynthetic gene clusters highlights the untapped potential of deadwood microorganisms for novel secondary metabolite production. Conclusions Our study emphasizes new contributors to wood decomposition, especially Polyangiaceae and Patescibacteria for complex and easily decomposable organic matter, respectively. The identification of nitrogen-fixing capabilities within the Steroidobacteraceae family introduces novel perspectives on nitrogen cycling in deadwood. The diverse array of observed biosynthetic gene clusters suggests intricate interactions among deadwood bacteria and promises the discovery of bioactive compounds. Long read sequencing not only advances our understanding of deadwood microbial communities but also demonstrates previously undiscovered functional capacities of the deadwood microbiome. Its application opens promising avenues for future ecological and biotechnological exploration of microbiomes. Bacteria biosynthetic gene cluster carbohydrate-active enzyme deadwood decomposition fungi metagenome-assembled genome metagenomics nitrogen fixation Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introduction Free-living microorganisms play a crucial role in global nutrient cycling, and their diverse metabolic abilities hold potential solutions for environmental and health issues. However, understanding microbial functional features in natural habitats is challenging due to the complexity of environmental microbiomes and the limitations of cultivation methods. Shotgun metagenomics offers an alternative to culturing, providing insights into taxonomic composition and functional potential of microbial species through the construction of metagenome-assembled genomes (MAGs) (Lemos et al., 2021 ). Unfortunately, due to the limitations of assembly performance, MAGs from second-generation (short-read) sequencing suffer from incompleteness, high level of fragmentation and contamination. Consequently, functional inferences may be misleading, the assembly of complete operons or large genomic islands is mostly impossible and identification is unclear since discriminative genes such as the 16S rRNA are largely missing (Parks et al., 2017 ). Third-generation (long-read) sequencing allows much more reliable assembly and thus holds the promise for recovering high-quality MAGs or complete/circular prokaryotic genomes from complex microbiome samples (Bickhart et al., 2022 ; Sereika et al., 2022 ), enabling an accurate assessment of their ecological roles. The two main long-read platforms include the cost-affordable Oxford Nanopore, albeit with inferior accuracy (Brown et al., 2021 ; Sereika et al., 2022 ), and the PacBio platform, offering highly accurate consensus sequencing (HiFi) reads without the need for further polishing (Jiang et al., 2023 ; Kim et al., 2022 ). In this study, we demonstrate the added value of PacBio HiFi sequencing over short-read sequencing for characterizing dominant microbiome members in decomposing deadwood. Deadwood, characterized by high carbon and low nitrogen concentrations, offers a variety of niches for microorganisms and represents a hotspot of biodiversity (Seibold et al., 2021 ). Although deadwood degradation contributes to release of CO 2 into the atmosphere, its final degradation products also enrich soils with nutrients and recalcitrant organic matter (OM), assisting nutrient cycling and carbon storage in forest ecosystems (Baldrian et al., 2023 ). Here, we used deadwood samples from a natural mixed temperate forest in Central Europe, which were previously analyzed through bacterial genomics, metagenomics and metatranscriptomics using second generation high-throughput sequencing (Tláskal et al., 2021a , 2021b , 2017 ). These studies identified fungi and bacteria as the main contributors to deadwood microbiomes. Fungi are efficient lignin degraders (Boer et al., 2005 ) and play the primary role in deadwood decomposition, while some bacteria contribute to decomposition processes by degrading cellulose, hemicellulose and fungal biomass. Importantly, bacteria have an essential role in nitrogen cycling in the deadwood, and the unique ability of some taxa to fix atmospheric nitrogen (N 2 ) compensates the high C:N ratio typical for deadwood (Rinne-Garmston et al., 2019 ). Despite recent progress in understanding bacterial roles in deadwood, the information gained from genomes recovered from second-generation high-throughput sequencing remained incomplete and fragmented (Tláskal et al., 2021a , b ). Complete genomes could be recovered from bacterial isolates from deadwood, but culture bias limited this approach to cultivable phyla (Tláskal and Baldrian, 2021 ). Here, we show that PacBio HiFi sequencing of long DNA molecules overcomes these limitations, providing higher numbers of MAGs than the assembly of short reads, the MAGs being moreover more complete, less contaminated and less fragmented. Assembly of DNA contigs containing complete operons with crucial functions allow better evaluation of the functional potential of deadwood bacteria. Ultimately, long read metagenomics in combination with metatranscriptomics identified new key actors in wood decomposition that remained “hidden” for the more traditional approaches. Material and Methods Sample description Deadwood samples were obtained from the core zone of the Žofínský prales National Nature Reserve, Czech Republic (48°39′57″N, 14°42′24″E). The core zone of the Žofínský prales is an unmanaged forest without any human interventions since 1838. The site and deadwood sampling were previously described in detail (Tláskal et al., 2021a ). A map generated with the r package sf (Pebesma, 2018 ) representing the location of the sampling site is provided in Fig. 1 . For this study, four samples from Fagus sylvatica trunks representing class 2 (4–7 years of decomposition; sample 6 – BioSample accession SAMN13925154 and sample 7 – SAMN13925155) and class 4 (20–41 years of decomposition; sample 57 – SAMN13925167 and sample 84 – SAMN13925168) were chosen. The samples were stored at -80°C prior to DNA isolation. Short-read sequencing using Illumina HiSeq platform was previously performed on these samples (Tláskal et al., 2021a , b ) and the short-read data were used in this study. DNA isolation High molecular weight DNA was isolated from the four deadwood samples using a phenol/chloroform/isoamyl alcohol extraction method according to Sagova-Mareckova et al. ( 2008 ) with several modifications; notably we replaced the bead-beating step by vortexing in order to prevent excessive DNA shearing. Samples were first thoroughly ground in a sterile mortar with liquid nitrogen using a rough pestle. Sample aliquots (~ 250 mg) were added to 2 mL tubes with a screw lid and mixed with 600 µL extraction buffer (50 mM Na-phosphate buffer pH 8, 50 mM NaCl, 500 mM Tris-HCl pH 8, and 5% sodium dodecyl sulfate) and 300 µL phenol (pH 8)/chloroform/isoamyl alcohol (25:24:1). The tubes were vortexed for 1 min at 4°C, using the Vortex-Genie (Scientific Industries, Bohemia, NY) set at the maximum speed. The homogenized samples were centrifuged at 10,000 g for 3 min and supernatant was transferred to a clean tube. One volume of phenol/chloroform/isoamyl alcohol (25:24:1) was added and mixed with the supernatant by gentle tube inverting for 1 min. After centrifugation at 6,000 g for 5 min, the supernatant was transferred to a clean tube, mixed (gentle tube inverting for 1 min) with 1 volume of chloroform/isoamyl alcohol (24:1), and centrifuged (6,000 g, 5 min). The supernatant was transferred to a clean tube and mixed with NaCl (to a final concentration of 1.5 M) and CTAB (to a final concentration of 1%), and incubated at 65°C for 35 min. The incubated solution was cooled at 4°C for 5 min and then mixed with an equal volume of chloroform/isoamyl alcohol (24:1), and centrifuged at 3,400 g for 20 min. The supernatant was mixed with 0.6 volume of isopropanol and 0.1 volume of 3M sodium acetate in a clean tube and incubated at 4°C for 45 min to precipitate the DNA. After centrifugation at 10,000 g for 20 min, the supernatant was removed and the DNA pellet was washed with 200 µL cold 70% ethanol, air-dried and resuspended in 30 µL 10 mM Tris buffer pH 8. Presence of high molecular weight DNA was verified with DNA electrophoresis in 0.8% agarose gel and DNA concentration was measured with Qubit 2.0 Fluorometer (Invitrogen, Carlsbad, CA). Three to six DNA aliquots per sample were pooled to obtain the minimal required DNA quantity of 5 µg. Sequencing and primary analyses of the metagenome The library preparation and PacBio HiFi sequencing on the Sequel II Instrument were performed at Brigham Young University Sequencing Centre, Utah. Samples 6 and 7 were pooled equimolarly during library preparation and sequenced together on one 8M SMRT Cell using 30-h movie. Samples 57 and 84 were sequenced separately on one 8M SMRT Cell using 30-h movie. PacBio’s movie subreads files were processed with CCS ( https://github.com/PacificBiosciences/pbbioconda ) for consensus Hi-Fi reads of > 99% accuracy. After removing reads of less than 1,000 bp, a total of 16.1×10 9 bp were obtained for the 4 samples ( Table S1 ). Next, taxonomic profiling of the raw PacBio HiFi reads was performed using DIAMOND + MEGAN-LR workflow, where reads are aligned with DIAMOND v2.1.8 (Buchfink et al., 2021 ) blastx against the NCBI nr database (version 09/2022) using the following parameter ‘--range-culling --top 5 -F 5000’ for long-read mode, followed by ‘meganization’ where LCA (last common ancestor) of each read is assigned. For this, the MEGAN6 Community Edition (Huson et al., 2016 ) ‘daa2rma’ function was used with the following parameter ‘--longReads --lcaAlgorithm longReads -ram readCount’. The metagenome and metatranscriptome of these samples previously sequenced on an Illumina HiSeq 2500 (2 × 250 bases), were downloaded from Tláskal et al. ( 2021b ). The quality of the reads was checked using Trimmomatic (v0.36) (Bolger et al., 2014 ) and FASTX-Toolkit ( http://hannonlab.cshl.edu/fastx_toolkit/ ), removing the adaptor contamination, the low quality reads (< 30), the reads < 50bp and trimming the low-quality ends of reads. Only the metatranscriptomes of samples 6 and 7 were available, containing 13×10 6 and 71×10 6 sequences, respectively ( Table S1 ). The Illumina HiSeq sequencing of the metagenomes generated a total of 19.8 Gb. Taxonomic profiling of the raw Illumina short-read was performed using Kraken2 v2.1.2 (Wood et al., 2019 ) against the NCBI nt database using default parameters. Taxonomic profiles of raw short reads suffered from a high share (~ 30%) of unassigned reads compared to raw long reads ( Table S2 ). Sequence assembly and taxonomic profiling Sequences generated by PacBio HiFi sequencing were first co-assembled (from the four samples) using Hifiasm-meta r058 (Feng et al., 2022 ) with default parameters, and metaFlye v2.9.2 (Kolmogorov et al., 2020 ) with --pacbio-hifi and --meta setting (Kim et al., 2022 ). Hifiasm-meta produced a total of 16,446 contigs with an N50 of 80,906bp, while metaFlye produced a total of 6,589 contigs with a N50 of 90,042bp ( Table S3 ). As a result, only Hifiasm-meta co-assemblies were further analyzed and Hifiasm-meta assembler was also used for sample-by-sample assembly. Sequences generated by Illumina HiSeq were co-assembled and sample-by-sample assembled using megahit v1.2.9 (Li et al., 2015 ) with default settings. Hybrid assemblies were generated using Unicycler v0.5.0 (Wick et al., 2017 ) with default settings on samples 57, 6 and 7 only, as the PacBio HiFi sequencing depth of sample 84 was too low, extremely increasing runtime. The hybrid assembly of each sample was computationally more demanding than those produced with megahit and hifiasm-meta. Furthermore, as the hybrid assemblies yielded contigs shorter in length compared to PacBio HiFi and in fewer numbers than Illumina ( Table S5 ), hybrid assembly was not further investigated. Taxonomy composition of assembled contigs was based on predicted genes, by FragGeneScan v1.31 (Rho et al., 2010 ), for co-assemblies and sample-by-sample assemblies. Predicted genes were aligned to NCBI nr (version 09/2022) database with DIAMOND blastp option, and taxonomy was based on best-hit (lowest e-value, highest identity and bit-score), with a minimum e-values of 10E-5. MAG recovery and taxonomic identification MAGs were generated using the sample-by-sample assembly and the co-assembly of short and long reads. We employed three binning tools (MetaBAT v2.12.1, MaxBin v2.2.6, and CONCOCT v1.0.0) with default settings within metaWRAP v1.3.2 (Uritskiy et al., 2018 ). The resulting bins were then combined and refined using the Bin_refinement module of metaWRAP with the parameters ‘-c 50 -x 10’ to obtain at least medium-quality draft MAGs. CheckM v1.0.12 (Parks et al., 2015 ) lineage workflow in metaWRAP was used to assess the quality (completeness and contamination) of the MAGs and seqkit v2.3.0 (Shen et al., 2016 ) was used for basic MAG stats. The MAGs were dereplicated using dRep v3.4.0 (Olm et al., 2017 ). Primary clusters (~ 90% ANI) were formed using Mash and secondary clusters (95% ANI) were created by aligning the genomes via ANImf. Ultimately, we obtained a total of 69 species-level MAGs from PacBio HiFi long read assemblies. We used GTDB-Tk v2.1.0 (Chaumeil et al., 2019 ) to assign taxonomy to these MAGs based on the taxonomy R207_v2 from the Genome Taxonomy Database (GTDB). We re-estimated the completeness and the contamination of the Patescibacteria MAGs using the recently developed Checkm2 v0.1.3 (Chklovski et al., 2023 ), providing better estimates for this phylum. Co-assembly generated a total of 28 MAGs, with a median completeness of 65.4% and median contamination of 1.6%, including 5 of the 6 Patescibacteria and 5 of the 7 Gammaproteobacteria. Sample-by-sample assembly produced a total of 41 MAGs, with a median completeness of 71.5% and median contamination of 0.7%, including 5 of the 6 Verrucomicrobiota, 8 of the 10 Bacteroidota and the 2 Myxococcota ( Table S6 ). Functional annotation of predicted ORFs was done with DRAM v1.4.0 (Shaffer et al., 2020 ), which relied on Pfam, KOfam, UniProt, dbCAN, and MEROPS. We estimated the relative abundances of the 69 MAGs across the four metagenome datasets using the CoverM v0.6.1 (Aroney et al., 2024 ) for genome abundance estimation, by assessing the coverage of mapped long reads in ‘genome’ mode. Results expressed in relative abundance were calculated by dividing the average genome coverage by the average total coverage of all genomes, then normalized by the ratio of mapped reads to total reads. Further contamination assessment and chimerism check was done using GUNC v1.0.5 (Orakov et al., 2021 ). We screened for the presence of ribosomal RNAs (i.e., 16S, 23S and/or 5S rRNA) in MAGs using barrnap v0.9 ( https://github.com/tseemann/barrnap ). Integrating MAGs and metabarcoding data The prokaryotic diversity of the samples was estimated using 16S rRNA amplicon sequencing data and compared with the 16S rRNA profiles of MAGs. First, Illumina MiSeq (2 × 250 bases) metabarcoding data for these samples were obtained from Tláskal et al., ( 2021b ). Subsequently, the amplicon sequencing data were processed using the SEED v2.1.05 (Větrovský et al., 2018 ) pipeline. Paired-end reads were joined using fastq-join, and sequences with ambiguous bases or a mean quality score below 30 were excluded. Chimeric sequences were identified and removed using USEARCH v11.0.667 (Edgar, 2010 ), followed by clustering with UPARSE within USEARCH (Edgar, 2013 ) at 97% of similarity. The most abundant sequences were chosen as representatives for each Operational Taxonomic Unit (OTU), singletons were discarded, and OTU tables were rarefied to a common sampling depth of 1,583 OTUs/sample using the rarefy_even_depth() function in the phyloseq R package (McMurdie and Holmes, 2013 ). Species-level matches were determined by BLASTn against Silva v138.1 (Quast et al., 2012 ). Hill numbers were used to calculate diversity indexes (observed richness = H0 and exponential of Shannon = eH’) using the DivPart() function in the entropart R package (Marcon and Herault, 2015 ). Finally, the 16S rRNA sequences of the MAGs were aligned and mapped to the 16S rRNA amplicon sequences using blastn (maximum e-value of 1 × 10 − 5 ) and bowtie2 v2.4.1 (Langmead and Salzberg, 2012 ), respectively. Alignment and mapping results were consistent for 45 out of the 48 MAGs possessing a 16S rRNA copy in their genome ( Table S8 ), and in agreement with the MAG GTDB taxonomy. In addition, the relative abundance of MAGs and their corresponding OTUs was correlated (Fig. 2 D). Functional characterization of the deadwood microbial communities Relevant functions in wood decomposition were screened (Tláskal et al., 2021a ). All these analyses were performed on short reads, long reads and MAGs, and the results were compared. EukRep v0.6.7 (West et al., 2018 ) were used to predict contigs of eukaryotic origin. Metatranscriptomes were used to validate the expression of specific metabolisms. Characterization of CAZymes Genes involved in the decomposition of the easily degradable carbohydrates (arabinogalactanases, xylanases/xyloglucanases, mannanases, cellobiases and xylobiases), complex plant cell wall biopolymers (endoglucanase, cellobiohydrolase, exoglucanase), microbial biomass (peptidoglycansases, beta-glucanases, chitinases) and reserve compounds (alphaglucanases) were predicted using FragGeneScan v1.31 (Rho et al., 2010 ), and annotated using run_dbcan.py (v2) program (Zhang et al., 2018 ). Predicted prokaryotic and eukaryotic genes were compared to the dbCAN database V9 using HMMER 3.0 (Finn et al., 2011 ). See Table S11 for the details of the CAZymes screened for each activity. To validate Myxococcota cellulose decomposition activity, we mapped transcripts from sample 7 to Myxococcota MAGs (both Myxococcota genomes were assembled from this sample) using minimap2 v2.24 (Li, 2021 ) with the ‘-x sr’ setting. Transcript counts were summed using dirseq v0.4.3 (Woodcroft et al., 2018 ) based on gene coordination from the gene prediction. The number of CAZyme for targeted activity were proportional between the short reads and the long reads ( Table S9 ). Investigating the genomic regions of nitrogen fixation Predicted genes from long reads and short reads co-assemblies were first screened with HMMER v3.3.2 against the KOfam database (accessed in June 2022) using the predefined adaptive thresholds (Aramaki et al., 2020 ). The number of metabolic pathways involved in the nitrogen cycle found in a single contig is reported in Table S9 . We further analyzed the contigs in which genomic region associated with N 2 fixation (i.e. nif genes + fix genes + Isc system genes or Suf system genes) were detected. We performed gene prediction using GeneMarkS v1.14 (Besemer, 2001 ) with default setting, and the annotation using eggnog-mapper v2.1.11 (Cantalapiedra et al., 2021 ) and MMseqs after translating input CDS to proteins. Only contigs from PacBio HiFi co-assembly contained essential genes for nitrogen fixation, and five of them were considered as potentially functional ( Table S12 ). The phylogenetic relationships between the nifH sequences were analyzed using neighbor-joining trees generated using the maximum likelihood algorithm with 1,000 bootstrap iterations. First, multiple sequence alignment was generated using mafft v7.490 (Katoh et al., 2019 ) with the option ‘--maxiterate 1000’ and ‘--localpair’, then sequences were trimmed using BMGE v.2.0 (Criscuolo and Gribaldo, 2010 ) software and the BLOSUM30 matrix. Finally, the tree was computed using iqtree v2.2.0.3 (Nguyen et al., 2015 ) and the LG + C20 model and visualized using FigTree v1.4.4 ( http://tree.bio.ed.ac.uk/software/figtree/ ). One of the nitrogen fixation contig (s509.ctg000513l) was also found in the 36_Steroidobacteraceae MAG, a gammaproteobacterial genome assembled in this study. To investigate nitrogen fixation potential in the Steroidobacteraceae family, we investigated 71 Steroidobacteraceae genomes from the GEMs catalog ( https://portal.nersc.gov/GEM/ ) together with 110 isolated strain genomes of Steroidobacteraceae from NCBI. We annotated these genomes using eggnog-mapper v2.1.11, with the option ‘--itype metagenome’ for gene prediction from Diamond/MMseqs2 blastx hits, and no nitrogenase genes were identified ( Table S12 ). Given the transferability of the essential genes for nitrogen fixation (Bolhuis et al., 2010 ), we studied genomic islands of 36_Steroidobacteraceae with Islandviewer4 (Bertelli et al., 2017 ) after having identified the start position of the genome with circlator v1.5.5 (Hunt et al., 2015 ), and plotted the results using Proksee (Grant et al., 2023 ). We found no evidence of nitrogen fixation capability acquired from horizontal gene transfer (HGT) in 36_Steroidobacteraceae. The genomic islands corresponded well to variation of GC content (Fig. 4 C), but occurred neither inside nor around the s509.ctg000513l contig (located between 5118742 and 5335241 bp). We further studied phylogenetic similarities of 36_Steroidobacteraceae nifH sequences in the Interpro database (accessed on August 1st, 2023). First, the nifH sequences were clustered at 97% of similarity using USEARCH v11.0.667. Then, multiple sequence alignment was generated using mafft v7.490 with the option ‘--maxiterate 1000’ and ‘--localpair’. We trimmed the alignment using BMGE v.2.0 software and the BLOSUM30 matrix. We computed the tree using iqtree v2.2.0.3 and the LG + C20 model and rooted the tree using minimal ancestor deviation (Tria et al., 2017 ). We next sub-selected the closest related taxa and used FigTree v1.4.4 for visualization. Finally, we confirmed the transcription of the nitrogen fixation genes in 36_Steroidobacteraceae by mapping the sample 7 transcripts (36_Steroidobacteraceae genomes were assembled from this sample) to s509.ctg000513l contig using minimap2 v2.24 with ‘-x sr’ setting. Transcript counts were summed using dirseq v0.4.3 (Woodcroft et al., 2018 ) based on gene coordination from the gene prediction. We also confirmed the nitrogen fixation gene expression in the five PacBio HiFi contigs by mapping the sample 7 transcript to nitrogenase genes ( nifH , nifD , nifK ) to make the results comparable ( Table S12 ). Biosynthetic Gene Clusters AntiSMASH v6.1.1 (Blin et al., 2021 ) was used to predict secondary metabolite clusters. BGCs were predicted with ‘--hmmdetection-strictness strict’ and ‘--asf’ settings. We also added the options ‘--cb-general‘, ‘--cb-subclusters‘ and ‘--cb-knownclusters‘ in order to compare identified clusters against antiSMASH-predicted clusters database, known subclusters responsible for synthesizing precursors and known gene clusters from the MIBiG database (Terlouw et al., 2023 ) respectively. To cluster the BGCs by similarity, we used BiG-SCAPE v1.1.5 (Navarro-Muñoz et al., 2020 ) with the ‘--mibig‘ and ‘--cutoffs 0.6‘ flags to cluster the BGC detected and the MIBiG database v3.1 entries at 60% percent of similarity. Results PacBio HiFi and Illumina sequence assembly and binning PacBio HiFi and Illumina HiSeq sequencing platforms generated respectively 16.1 Gb and 19.8 Gb for the four deadwood samples ( Table S1 ). For the samples 6, 7, and 57, PacBio HiFi sequencing generated on average 282,091 sequences per sample, with an average sum length of 5.2 Gb. For the sample 84, PacBio HiFi sequencing produced 19,734 sequences with an average sum length of 0.3 Gb. Illumina sequencing yielded on average 21,168,790 sequences per sample, with an average sum length of 4.9 Gb ( Table S1 ). The average sequence length from PacBio HiFi was approximately 18,300 bp and 234 bp from Illumina. Sequences affiliated to Prokaryotes (almost exclusively Bacteria) dominated in both PacBio HiFi raw reads (85%) and Illumina reads (51%), while the share of sequences affiliated to Eukaryotes was 13% for PacBio HiFi raw reads and 19% for Illumina raw reads ( Table S2 ). About 30% Illumina raw reads remained unclassified. The PacBio HiFi sequences were co-assembled into 16,446 contigs with an N50 of 66,188 bp, with 16,439 contigs > 10 kb and 9,696 contigs > 50 kb ( Table S3 ). The largest contig reached a size of 5.8 Mb. The Illumina short reads co-assembly generated 3,734,801 contigs of which 102,279 contigs were > 1 kb, and 109 were > 10 kb, with an N50 of 685 bp ( Table S3 ). The largest contig had a size of 0.1 Mb. The total length of assembled long-read contigs was 1.2 Gb, while for short-read contigs was around half that (i.e. 0.6 Gb). The average proportion of long and short read mapping to contigs was 41% and 39%, respectively ( Table S4 ). See Table S5 for details on sample-by-sample assembly for PacBio HiFi and Illumina HiSeq. Contigs affiliated to Prokaryotes dominated sample-by-sample assemblies and co-assemblies from both PacBio HiFi and Illumina data. Proportions of contigs affiliated to eukaryota were consistent between sample-by-sample assembly and co-assembly, accounting respectively for 5.5% and 5.1% of short reads and 8.5% and 10.5% of long reads ( Table S2 ). K-mer-based analysis supported these results, identifying 3,168 eukaryotic contigs (270 kb) in the PacBio HiFi co-assembly and 454 eukaryotic contigs (2 kb) in the PacBio HiFi co-assembly. Twenty one percent of the raw PacBio HiFi reads were successfully binned ( Table S6 ), constituting a total of 69 unique bacterial metagenome-assembled genomes (MAGs). Among these, 14 were High-quality draft MAGs (i.e. completeness > 90% and contamination 50%, contamination < 10%, Figure S1 ). Twenty-two MAGs were composed of less than 10 contigs, among which 7 were composed of a single contig and considered nearly finished MAGs ( Table S6 ). Ribosomal RNA (i.e., 16S, 23S and/or 5S rRNA) genes were found in 53 of the 69 MAGs ( Table S6 ). Eleven unique MAGs were recovered from the Illumina HiSeq co-assembly and sample-by-sample assembly, all of them Medium-quality draft MAGs ( Table S7, Figure S1 ). These MAGs consisted of 772 ± 603 contigs, with an average length of 3,480 ± 4,254 bp, while the MAGs generated from PacBio HiFi assemblies consisted of 26 ± 21 contigs with an average length of 148,129 ± 386,395 bp. Six of the 11 MAGs obtained from the short-read assemblies were also recovered from long-read data. Integrating de novo genome assembly and metabarcoding Based on metabarcoding, prokaryotic 16S sequences in the deadwood were most frequently assigned to Proteobacteria (43%), and to a lesser extent to Acidobacteriota (14%), Actinobacteriota (10%), Bacteroidota (9%), Verrucomicrobiota (9%), Planctomycetota (8%), Myxococcota (3%) and Patescibacteria (1%) (Fig. 2 A). MAGs from seven of these eight bacterial phyla with a relative abundance > 1% identified by 16S metabarcoding were successfully binned from PacBio HiFi assemblies. This included 38 genomes belonging to Proteobacteria (31 to Alphaproteobacteria and 7 to Gammaproteobacteria), 10 to Bacteroidota, 6 to Verrucomicrobiota, 6 to Patescibacteria, 4 to Acidobacteriota, 3 to Actinobacteriota, and 2 to Myxococcota (Fig. 2 B). The number of MAGs recovered per phylum showed positive correlation with the OTU richness of this phylum (Pearson correlation coefficient, cor = 0.84, p-value = 0.008). However, despite Planctomycetota contigs were recovered ( Figure S2 ), no Planctomycetota genomes were assembled, probably due to the specific community structure of this clade. Planctomycetota exhibited a relatively low 16S rRNA abundance compared to the seven other most abundant phyla (ranking as the sixth most abundant phylum) but were the second prokaryotic phylum after Proteobacteria in terms of OTU richness and diversity (Fig. 2 A). Among the 48 MAGs containing a copy of the 16S rRNA gene, 44 MAGs were successfully linked to a unique OTU recovered using 16S metabarcoding. In the remaining four cases, two different MAGs aligned with a single OTU ( Table S8 ), likely due to a low resolution (e.g. size) of the 16S rRNA gene fragment. Within these 48 MAGs, 24 were associated with OTUs exhibiting a relative abundance of 16S greater than 0.1% (i.e., among the 172 OTUs, Fig. 2 C), including 5 MAGs originating from the 12 most abundant OTUs. Of these, 3 MAGs corresponded to Alphaproteobacteria and 2 corresponded to Myxoccocota. Additionally, 20 MAGs corresponded to OTUs with a relative abundance, based on metabarcoding, below 0.1%. This included two assigned to Patescibacteria and two assigned to Alphaproteobacteria. While the relative abundance of the MAGs where generally proportional to the relative abundance of the respective OTU (Fig. 2 D), one abundant Patescibacteria MAG (5_UBA9983_A) had a low 16S relative abundance ( Table S6 , Table S8 ). CAZymes encoded by the deadwood microorganisms The abundance of CAZymes for each activity was comparable across short reads, long reads, and MAGs ( Figure S3 ). In the short-read co-assembly, we identified a total of 7,617 CAZymes, while 5,161 CAZymes were found in the long-read co-assembly. Among these, alphaglucanases, peptidoglycanases, cellobiases, and xylobiases were the most abundant enzymes, whereas exoglucanases and cellobiohydrolases were relatively rare. Notably, the majority of identified CAZymes were bacterial in origin. Short-read data did not yield any eukaryotic CAZymes, whereas 197 eukaryotic CAZymes were identified in the PacBio HiFi long-read data. When considering the total gene count in the long-read co-assembly, the proportion of CAZymes represented 0.37% for bacteria and 0.04% for eukaryotes ( Table S9 ). The decomposition of easily decomposable OM mediated by cellobiases and xylobiases was frequently observed in the recovered genomes (Fig. 3 ), with only 3 MAGs lacking this capability. Additionally, CAZymes targeting reserve compounds and microbial biomass were abundant across most MAGs, although relatively rare in Patescibacteria and a few other MAGs. Notably, cellulose and chitin decomposition capabilities were less common, present in 52% and 51% of the genomes, respectively. MAGs affiliated with Myxococcota (Polyangiaceae family) encoded the highest number of CAZymes involved in the degradation of both readily decomposable and recalcitrant biopolymers, including those targeting reserve compounds. However, they lacked enzymes targeting fungal biomass decomposition (chitinases). Importantly, Myxococcota MAGs exhibited 14 times more cellulose-targeting enzymes than other MAGs in this study and possessed the set of all necessary enzymes for cellulose decomposition (endoglucanases, exoglucanases, and cellobiohydrolases). Transcript mapping revealed that endoglucanases were commonly transcribed by all prokaryotic MAGs where they were present ( Table S10 ). However, exoglucanases were predominantly transcribed by Polyangiaceae (Myxococcota) MAGs, and cellobiohydrolases were exclusively transcribed by these MAGs. Bacteroidetes exhibited a high proportion of CAZymes targeting microbial biomass, along with numerous enzymes targeting easily degradable carbohydrates and a few targeting reserve compounds. Conversely, Acidobacteriota and Verrucomicrobiota showed proportionally fewer enzymes targeting microbial biomass but many involved in the decomposition of easily degradable carbohydrates and reserve compounds. Proteobacteria encoded numerous CAZymes targeting microbial biomass, with an uneven distribution of enzymes involved in easily degradable carbohydrates and reserve compound decomposition, with some encoding few and others encoding many ( Table S10 ). The number of CAZymes found in MAGs was proportional to their genome size (cor = 0.76, p-value < 0.001). Consequently, the phyla encoding the smallest number of CAZymes were Actinobacteriota and Patescibacteria. Actinobacteriota assembled in deadwood (Nanopelagicaceae and Microbacteriaceae families), encoded mainly CAZymes targeting reserve compounds, had limited number of CAZymes targeting easily degradable carbohydrates, and lacked the metabolic capacity to degrade complex biopolymers of plant cell wall and microbial biomass (Fig. 3 ). Patescibacteria exhibited only a few enzymes involved in decomposition of easily degradable carbohydrates, targeting microbial biomass and reserve compounds. However, transcript mapping confirmed expression of CAZymes by Patescibacteria, exceeding the transcription of Acidobacteriota and Proteobacteria for mannanase and betaglucanases activity ( Table S10 ). Importantly, while the level of chitinase expression was generally low among recovered MAGs (mean coverage = 3.0), Patescibacteria were the second most active phylum in terms of chitinase expression after Bacteroidota (6.6 and 7.3 respectively). Nitrogen cycling metabolism The number of metabolic pathways involved in the nitrogen cycle was low. Of those identified, dissimilatory and assimilatory nitrate reduction were the most frequent in Pacbio HiFi contigs, followed by nitrogen fixation (Table S9) . In addition, a contig with the ammonia methane/monooxygenase ( pmo - amo ) ABC subunit was identified, suggesting a potential for nitrification. Consistently, genes coding for N 2 fixation and dissimilatory or assimilatory nitrate reduction were present in 14 MAGs (Fig. 3 ). The genes nirB and nirD , converting nitrite to ammonia in the second step of dissimilatory nitrate reduction pathway, were found in Proteobacteria (3 Alphaproteobacteria MAGs), Myxococcota (2 Polyangiales MAGs) and Acidobacteriota (Bryobacterales). The gene nirA , catalyzing the conversion of nitrite to ammonia in the second step of assimilatory nitrate reduction pathway, was found in Verrucomicrobiota (Methylacidiphilales) and Alphaproteobacteria (3 Rhizobiales MAGs). The genes nifH , nifD , and nifK (i.e. nitrogenase genes), responsible for the biological reduction of dinitrogen to ammonia, were only found in one Gammaproteobacteria MAG belonging to the Steroidobacterales. To gain deeper insights into nitrogen fixation during wood decomposition processes, we examined the genomic regions associated with N 2 fixation in both short-read and long-read contigs. No contigs generated by the short-read co-assembly contained essential genes for nitrogen fixation, whereas the PacBio HiFi co-assembly revealed five contigs with genomic regions associated to N 2 fixation (Fig. 4 A). All nitrogenase genes identified in these contigs were transcribed (Fig. 4 B), suggesting that they are functional during in situ nitrogen fixation. The minimum set of genes encoding structural and biosynthetic components — specifically NifHDK and NifENB — were present. The fixABCRUX genes were also present, preferred to the Rnf complex to catalyze the production of reduced ferredoxin/flavodoxin. Four of these contigs were affiliated to Alphaproteobacteria and one to Gammaproteobacteria ( Table S12 ). The Gammaproteobacteria contig lacked nifS but encoded the complete Isc system ( iscR , iscS , iscU , iscA , hscB , hscA , fdx , and iscX ), required for maturation of [Fe-S] proteins. This system was absent in the Alphaproteobacteria contigs, where it was substituted by the Suf system ( sufB , sufC , sufD, sufS , and sufE ). The gammaproteobacterial nitrogen fixation contig (s509.ctg000513l) was binned into the 36_Steroidobacteraceae MAG. Transcript mapping against this MAG validated the expression of essential genes for nitrogen fixation ( Table S12 ). The identification of nitrogen fixation genes within the Steroidobacteraceae family is noteworthy. Despite annotating 181 Steroidobacteraceae genomes from the Earth's Microbiomes catalog and the NCBI database (71 and 110 respectively, Table S13 ), revealing individual genes potentially implicated in nitrogen fixation (e.g., fixB , iscASX , nifZ , sufBCES , fdx ), none of the annotated genomes encompassed the complete set of essential nitrogen fixation genes ( Table S12 ). Moreover, extensive investigation of the NCBI nr and Interpro databases ( Table S12 ) failed to identify nifH in the Steroidobacteraceae family. Phylogenetic analyses of 36_Steroidobacteraceae nifH and the entire s509.ctg000513l contig indicated similarities with the Methyloccocales order (Fig. 4 D, Table S12 ), yet we found no evidence supporting recent HGT of this region into the Steroidobacteriaceae MAG (Fig. 4 C, see Methods for details). Our analyses also found no support for mis-binning, affirming the accurate assembly of 36_Steroidobacteraceae ( Table S6 ). Collectively, these data underscore a functional potential for nitrogen fixation within this deadwood-associated member of the Steroidobacteraceae family. Biosynthetic Gene Clusters in deadwood microorganisms A total of 1,089 putative Biosynthetic Gene Clusters (BGCs) were identified in the PacBio HiFi long-read co-assembly, presenting a stark contrast to the three BGCs found in the short-read co-assembly ( Figure S4 ). The most frequent BGC types were terpenes (291) and ribosomally synthesized and post-translationally modified peptides (RiPPs, 235). While the majority of BGCs were located on bacterial contigs, we also uncovered eukaryotic BGCs, including terpenes, non-ribosomally synthesized peptides (NRPS), polyketides, and siderophores ( Figure S4 ). A comparison with the MIBig database highlighted the novelty of the identified BGCs, with a staggering 98% showing less than 50% shared genes with MIBig entries. The primary BGC types identified in the long-read assembly were also found in MAGs ( Figure S4 ), enabling the association of biosynthetic potential with taxonomic groups. Proteobacteria, particularly MAGs from Rhizomicrobium, Rhizobiales (Alphaproteobacteria), and Steroidibacteriaceae (Gammaproteobacteria), carried the majority of BGCs, exceeding 10 in some instances (Fig. 5 A). Proteobacterial BGCs presented a wide diversity, encompassing various types of compounds such as thioamitides, lassopeptides, ranthipeptides, azol(in)e-containing linear peptides, linaridines, or ladderanes, each holding bioprospecting potential. The two Myxococcota MAGs displayed the highest numbers of BGCs (14 and 15), encoding a wide array of compounds, including thioamitides and ranthipeptides. Verrucomicrobiota, Acidobacteriota, Bacteroidota, and Actinobacteriota MAGs harbored 0–8 BGCs, with terpene, RiPP, polyketide, NRPS, and arylpolyene BGCs being the most prevalent. Group-specific BGCs, such as flexirubins in Chitinophagaceae (Bacteroidota) MAGs, were also identified. Patescibacteria MAGs did not contain any BGCs. BiG-SCAPE analysis revealed that the 271 BGCs identified in MAGs were categorized into 254 unique gene cluster families (Fig. 5 B). The abundance of singletons underscored the vast diversity of BGCs in deadwood MAGs. For instance, seven BGCs encoding thioamitides each belonged to a distinct gene cluster family (Fig. 5 C). BGCs belonging to one cluster family always belonged to taxonomically related MAGs, indicating that BGCs in deadwood bacteria are phylogenetically conserved. Discussion In this study, we compared the performance of PacBio HiFi and Illumina sequencing platforms and tested different methods of assembly for metagenomes of deadwood samples. The two platforms generated roughly similar amounts of data, but the PacBio HiFi sequencing produced much longer reads than Illumina. Hifiasm-meta produced the best assemblies for PacBio HiFi data, yielding twice the amount of data and longer contigs than Illumina megahit assemblies. Hybrid assembly was more computationally demanding, and did not improve the number and size of contigs. This is probably because hybrid assembly was originally developed to improve short-read assembly (Wick et al., 2017 ), as the sequencing error generated by long-read sequencing was high. However, the repetitive library which calls for consensus reads developed by the PacBio HiFi sequencing approach, has significantly improved nucleotide accuracy (Marx, 2023 ). As a result, short reads are no longer needed to improve the assembly of long reads if sequenced by PacBio HiFi. PacBio HiFi metagenomic sequencing has demonstrated outstanding efficiency for reconstructing the genomes from deadwood microbial communities, including bacteria that could not be previously cultivated from these samples (Tláskal et al, 2021) or elsewhere. A total of 69 bacterial genomes were generated from 16 Gb of PacBio HiFi reads from 4 deadwood samples, including 67 newly reconstructed MAGs, 14 high-quality and 7 composed of a single contig. It outperformed Illumina assemblies by 6.3 times (11 MAGs), producing less fragmented, less contaminated and more complete genomes. In comparison to related studies, our findings overtake those of (Tláskal et al., 2021b ), who assembled 58 MAGs from 25 short-read metagenomes of deadwood from the same forest. Our investigation extended to the eight most abundant deadwood bacterial phyla, revealing successful genome assembly for all phyla except Planctomycetota. Our results suggest that while genome size and relative abundance may not be the ultimate limiting factors for successful binning, the inherent complexity of Planctomycetota, characterized by high species richness and phylogenetic diversity (Fig. 2 A), challenges the recovery of their genomes. Illumina and PacBio HiFi sequencing generated a proportional number of eukaryotic sequences, but PacBio HiFi assemblies yielded more eukaryotic contigs than Illumina assemblies. This discrepancy is likely attributed to the longer sequencing lengths and high accuracy of PacBio HiFi sequencing (Uliano-Silva et al., 2023 ). Eukaryotic genomes, characterized by intricate features like repetitive regions, introns, and exons (Galagan et al., 2005 ), require greater metagenomic sequencing depth and sample size to be successfully assembled compared to prokaryotic genomes. For instance, (Saraiva et al., 2023 ) needed 6,000 terrestrial metagenomes to assemble 197 eukaryotic bins, whereas (Ma et al., 2023 ) achieved the assembly of 40,039 prokaryotic MAGs from 2,941 soil metagenomes. Employing PacBio HiFi sequencing, we unveiled the important role of Myxococcota in deadwood decomposition. While Myxococcota have been previously assembled from various ecosystems, including aquatic, terrestrial, host-associated, and built environments (Nayfach et al., 2021), our study marks the first assembly of Myxococcota genomes within the deadwood ecosystem. Characterized by relatively large genomes (8 Mbp according to GTDB database r202), Myxococcota exhibit unique traits, encompassing motility (Nan et al., 2013 ), predation (Thiery and Kaimer, 2020 ), fruiting bodies formation (Muñoz-Dorado et al., 2016 ), and the ability to decompose cellulose (López-Mondéjar et al., 2022 ). Here, we successfully assembled two abundant Myxococcota (Polyangiaceae) genomes expressing high cellulosic activity, including endoglucanase, exoglucanase, and cellobiohydrolase – enzymes exclusively found in Myxococcota MAGs ( Table S10 ). While Polyangiaceae also expressed CAZymes targeting bacterial biomass, the lack of chitinase prevented them from recycling fungal biomass. Conversely, chitinases were consistently present in Bacteroidota and detected in Verrucomicrobiota, Proteobacteria and Patescibacteria. Remarkably, the sole chitinase found in Patescibacteria 5_UBA9983_A demonstrated the second-highest level of expression ( Table S10 ). Patescibacteria had been identified in a wide range of ecosystems (Nayfach et al., 2021) including deadwood habitats (Choi et al., 2022 ; Tláskal et al., 2021b ), yet their ecological role remains enigmatic. While their limited genome size, lacking crucial metabolic genes, suggests an obligate epibiotic lifestyle (Kuroda et al., 2022 ), their parasitic status remains subject to debate (Wang et al., 2023 ). In our study, UBA9983_A demonstrated a specialized opportunistic lifestyle, exclusively recycling fungal cell wall components ( Table S10 ). Additionally, Patescibacteria of the order Sacharrimonadales showcased elevated mannase expression, an enzyme involved in hemicellulose degradation, highlighting their active contribution in wood decomposition processes. These findings collectively imply that Patescibacteria may not solely depend on host resources. Further analyses are needed to elucidate whether hemicellulose- and fungal-derived residues serve as energy sources for ATP generation or benefits to the host (e.g., commensalism or mutualism). Nevertheless, our results suggest a broader ecological relationship for Patescibacteria beyond parasitism. Nitrogen concentration in deadwood is low (Tláskal et al., 2021a ) and probably represents a limiting factor for efficient decomposition. As metabolic pathways involved in nitrogen assimilation were scarce in the studied samples ( see Results ), microbial biomass could represent an alternative source of nitrogen (López-Mondéjar et al., 2018 ). We found that CAZymes targeting bacterial biomass were more frequent than CAZymes targeting fungal biomass, but that peptidoglycanases were 2 times less expressed than chitinases ( Table S11 , Table S10 ). This result is surprising since fungal biomass contains less nitrogen than bacterial biomass (Paul and Frey, 2024 ), but these processes might simply be regulated by the availability of bacterial and eukaryotic biomass. While microbial biomass decomposition represents an interesting strategy for recycling nitrogen during decay, an external source of nitrogen might be required to initiate decomposition of fresh dead wood. The initial nitrogen input is likely provided by bacterial nitrogen fixation, as illustrated by nitrogen fixation rates being eight times higher in young deadwood compared to old deadwood (Tláskal et al., 2021a ). The conversion of atmospheric N 2 to NH 3 , being a highly energy-consuming process (Cherkasov et al., 2015 ), is only realized if no better suitable nitrogen sources are available (Burris and Roberts, 1993 ). The continuous expression of nitrogenase genes in later stages of decomposition (> 4 years) thus indicates that the recycling of microbial biomass does not completely meet the microbial nitrogen demand throughout the decomposition process. Despite biological nitrogen fixation has been extensively investigated (e.g. Davies-Barnard and Friedlingstein, 2020 ; Dos Santos et al., 2012 ; Zehr and Capone, 2020 ), our study marks the first recovery of this function within the Steroidobacteraceae family. Beyond reporting a novel nitrogen-supplying bacteria in a nitrogen-limited environment, PacBio HiFi sequencing facilitated the investigation of genes involved in the conversion of N 2 into ammonium. In the deadwood ecosystem, Proteobacteria catalyze the production of reduced ferredoxin/flavodoxin using the fix operon, particularly advantageous under oxygen-limited conditions (Alleman et al., 2022 ). In addition, Steroidobacteraceae employ the isc system for biosynthesizing [Fe-S] proteins, crucial under elevated oxygen conditions (Johnson et al., 2006 ). The coexistence of the fix operon and the isc system likely allows Steroidobacteraceae to maintain nitrogen fixation activity under oxygen concentration fluctuations. Alphaproteobacteria appear less sensitive to elevated oxygen, favoring the suf system over the isc system, which is known to be more beneficial for bacterial growth in the presence of hydrogen peroxide (Tokumoto, 2004 ). Although the proportional contribution of Steroidobacteraceae (Gammaproteobacteria) and Alphaproteobacteria to global deadwood nitrogen fixation requires further exploration, our study provides empirical evidence of their in situ activity. PacBio HiFi sequencing unveiled the remarkable potential of deadwood microorganisms for the production of diverse secondary metabolites, an important trait that would be totally neglected if one would just rely on short read-based approaches. Our study identified over a thousand mostly novel BGCs, showcasing the extensive diversity of these BGCs in MAGs. This diversity suggests that deadwood bacteria display multiple interactions with other microorganisms in deadwood. Furthermore, our investigation highlighted bacterial groups in deadwood that hold promise for the production of novel bioactive compounds. Similar to observations in soil (Sharrar et al., 2020 ), the abundance of BGC types varied by taxonomy. In deadwood, we observed high numbers of BGCs in Myxococcota and Proteobacteria, as well as in other groups such as Verrucomicrobiota and Acidobacteriota, whose biosynthetic potential has only recently been reported (Crits-Christoph et al., 2018 ; Waschulin et al., 2022 ). Notably, we did not identify a shared BGC family across different bacterial taxa, indicative of a compound with broader importance in deadwood. In contrast, taxonomic conservation of BGC families in deadwood MAGs was observed. The presence of BGCs for flexirubins, pigments typical of Bacteroidota (Brinkmann et al., 2022 ), in several Chitinophagaceae MAGs constitutes convincing proof of validity of our metagenome analysis. Long-read sequencing-based analysis of metagenomes, while powerful, still has limitations. More effort would be needed to appropriately describe the fungal component of the deadwood microbiome and this is partly true also for the Planctomycetes: although their sequences have been obtained by PacBio HiFi sequencing, the high species richness and phylogenetic diversity of this phylum challenged their genome recovery. Still, long-read sequencing-based metagenome analysis stands unparalleled indicating microbiome functions not recoverable using the conventional approaches – short-read metagenomics and culturing. It enabled the assembly of novel genomes of bacteria with key roles in deadwood decomposition: cellulose decomposition in Polyangiaceae, nitrogen fixation in Steroidobacteraceae. It also revealed significant contribution of Patescibacteria to wood decomposition processes, and identified the wealth of new BGCs with potential ecological and/or biotechnological significance. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Availability of data and material Descriptions of the deadwood samples 6, 7, 57 and 84 are available at the NCBI BioSample repository (https://www.ncbi.nlm.nih.gov/biosample/), under accession numbers SAMN13925154, SAMN13925155, SAMN13925167, and SAMN13925168, respectively. The raw PacBio HiFi sequences from the four deadwood samples are available at the NCBI Sequence Read Archive repository (https://www.ncbi.nlm.nih.gov/sra/), under accession numbers SRR28211698 ̶ SRR28211701. The co-assembly of PacBio HiFi sequences is available at NCBI GenBank (https://www.ncbi.nlm.nih.gov/nuccore/) under accession number JBBCBH000000000. The 69 metagenome-assembled genomes from PacBio HiFi sequences are available at the NCBI GenBank repository, under accession numbers JBBCFX000000000 ̶ JBBCIN000000000. The raw Illumina HiSeq metagenome sequences corresponding to samples 6, 7, 57 and 84 are available at the NCBI Sequence Read Archive repository, under accession numbers SRR10968229, SRR10968228, SRR10968259, and SRR10968258, respectively. The Illumina HiSeq metatranscriptome sequences of samples 6 and 7 are available at the NCBI Sequence Read Archive repository, under accession numbers SRR10968251 and SRR10968250. The Illumina MiSeq 16S rRNA amplicon sequencing data for samples 6, 7, 57 and 84 are available at the NCBI Sequence Read Archive repository, under accession numbers SRR12914735, SRR12914734, SRR12914771, and SRR12914770. The biosynthetic gene clusters recovered from Illumina short read metagenome, PacBio HiFi long read metagenome and from PacBio HiFi MAGs, are available at the Zenodo repository (https://zenodo.org), under record numbers 10550953, 10529179, and 10529038, respectively. Competing interests The authors declare that they have no competing interests. Funding This work was supported by the Ministry of Education, Youth and Sports of the Czech Republic (CZ.02.01.01/00/22_008/0004635 - AdAgriF - Advanced methods of greenhouse gases emission reduction and sequestration in agriculture and forest landscape for climate change mitigation). Authors' contributions M.K. and P.B. designed the study. E.R., M.K. and P.B. directed the analyses. M.K. and V.T. performed HMW DNA isolation. P.T.D. and E.R. carried out the computational analyses and visual representations. E.R and M.K. analyzed the data with the help of P.T.D., V.T., R.L.M. and P.B. E.R and M.K. wrote the original manuscript, with the input of P.T.D., V.T., R.L.M. and P.B. All authors have read and approved the final manuscript. Acknowledgements The authors are grateful to the Nature Conservation Agency of the Czech Republic for granting the permit to perform environmental sampling within the Zofinsky prales National Natural Reserve. The authors acknowledge BioRender, which was used for figure creation. References Alleman, A.B., Garcia Costas, A., Mus, F., Peters, J.W., 2022. Rnf and Fix Have Specific Roles during Aerobic Nitrogen Fixation in Azotobacter vinelandii. Appl. Environ. Microbiol. 88, e01049-22. https://doi.org/10.1128/aem.01049-22 Aramaki, T., Blanc-Mathieu, R., Endo, H., Ohkubo, K., Kanehisa, M., Goto, S., Ogata, H., 2020. KofamKOALA: KEGG Ortholog assignment based on profile HMM and adaptive score threshold. Bioinformatics 36, 2251–2252. https://doi.org/10.1093/bioinformatics/btz859 Aroney, S.T.N., Newell, R.J.P., Nissen, J., Camargo, A.P., Tyson, G.W., Woodcroft, B.J., 2024. CoverM: Read coverage calculator for metagenomics. https://doi.org/10.5281/ZENODO.10531253 Baldrian, P., López-Mondéjar, R., Kohout, P., 2023. Forest microbiome and global change. Nat. Rev. Microbiol. 21, 487–501. https://doi.org/10.1038/s41579-023-00876-4 Bertelli, C., Laird, M.R., Williams, K.P., Simon Fraser University Research Computing Group, Lau, B.Y., Hoad, G., Winsor, G.L., Brinkman, F.S., 2017. IslandViewer 4: expanded prediction of genomic islands for larger-scale datasets. Nucleic Acids Res. 45, W30–W35. https://doi.org/10.1093/nar/gkx343 Besemer, J., 2001. GeneMarkS: a self-training method for prediction of gene starts in microbial genomes. Implications for finding sequence motifs in regulatory regions. Nucleic Acids Res. 29, 2607–2618. https://doi.org/10.1093/nar/29.12.2607 Bickhart, D.M., Kolmogorov, M., Tseng, E., Portik, D.M., Korobeynikov, A., Tolstoganov, I., Uritskiy, G., Liachko, I., Sullivan, S.T., Shin, S.B., Zorea, A., Andreu, V.P., Panke-Buisse, K., Medema, M.H., Mizrahi, I., Pevzner, P.A., Smith, T.P.L., 2022. Generating lineage-resolved, complete metagenome-assembled genomes from complex microbial communities. Nat. Biotechnol. 40, 711–719. https://doi.org/10.1038/s41587-021-01130-z Blin, K., Shaw, S., Kloosterman, A.M., Charlop-Powers, Z., van Wezel, G.P., Medema, M.H., Weber, T., 2021. antiSMASH 6.0: improving cluster detection and comparison capabilities. Nucleic Acids Res. 49, W29–W35. https://doi.org/10.1093/nar/gkab335 Boer, W. de, Folman, L.B., Summerbell, R.C., Boddy, L., 2005. Living in a fungal world: impact of fungi on soil bacterial niche development. FEMS Microbiol. Rev. 29, 795–811. https://doi.org/10.1016/j.femsre.2004.11.005 Bolger, A.M., Lohse, M., Usadel, B., 2014. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30, 2114–2120. https://doi.org/10.1093/bioinformatics/btu170 Bolhuis, H., Severin, I., Confurius-Guns, V., Wollenzien, U.I.A., Stal, L.J., 2010. Horizontal transfer of the nitrogen fixation gene cluster in the cyanobacterium Microcoleus chthonoplastes . ISME J. 4, 121–130. https://doi.org/10.1038/ismej.2009.99 Brinkmann, S., Kurz, M., Patras, M.A., Hartwig, C., Marner, M., Leis, B., Billion, A., Kleiner, Y., Bauer, A., Toti, L., Pöverlein, C., Hammann, P.E., Vilcinskas, A., Glaeser, J., Spohn, M., Schäberle, T.F., 2022. Genomic and Chemical Decryption of the Bacteroidetes Phylum for Its Potential to Biosynthesize Natural Products. Microbiol. Spectr. 10, e02479-21. https://doi.org/10.1128/spectrum.02479-21 Brown, C.L., Keenum, I.M., Dai, D., Zhang, L., Vikesland, P.J., Pruden, A., 2021. Critical evaluation of short, long, and hybrid assembly for contextual analysis of antibiotic resistance genes in complex environmental metagenomes. Sci. Rep. 11, 3753. https://doi.org/10.1038/s41598-021-83081-8 Buchfink, B., Reuter, K., Drost, H.-G., 2021. Sensitive protein alignments at tree-of-life scale using DIAMOND. Nat. Methods 18, 366–368. https://doi.org/10.1038/s41592-021-01101-x Burris, R.H., Roberts, G.P., 1993. Biological Nitrogen Fixation. Annu. Rev. Nutr. 13, 317–335. https://doi.org/10.1146/annurev.nu.13.070193.001533 Cantalapiedra, C.P., Hernández-Plaza, A., Letunic, I., Bork, P., Huerta-Cepas, J., 2021. eggNOG-mapper v2: Functional Annotation, Orthology Assignments, and Domain Prediction at the Metagenomic Scale. Mol. Biol. Evol. 38, 5825–5829. https://doi.org/10.1093/molbev/msab293 Chaumeil, P.-A., Mussig, A.J., Hugenholtz, P., Parks, D.H., 2019. GTDB-Tk: a toolkit to classify genomes with the Genome Taxonomy Database. Bioinformatics btz848. https://doi.org/10.1093/bioinformatics/btz848 Cherkasov, N., Ibhadon, A.O., Fitzpatrick, P., 2015. A review of the existing and alternative methods for greener nitrogen fixation. Chem. Eng. Process. Process Intensif. 90, 24–33. https://doi.org/10.1016/j.cep.2015.02.004 Chklovski, A., Parks, D.H., Woodcroft, B.J., Tyson, G.W., 2023. CheckM2: a rapid, scalable and accurate tool for assessing microbial genome quality using machine learning. Nat. Methods 20, 1203–1212. https://doi.org/10.1038/s41592-023-01940-w Choi, B.Y., Lee, S., Kim, J., Park, H., Kim, J.-H., Kim, M., Park, S.-J., Kim, K.-T., Ryu, H., Shim, D., 2022. Comparison of Endophytic and Epiphytic Microbial Communities in Surviving and Dead Korean Fir (Abies koreana) Using Metagenomic Sequencing. Forests 13, 1932. https://doi.org/10.3390/f13111932 Criscuolo, A., Gribaldo, S., 2010. BMGE (Block Mapping and Gathering with Entropy): a new software for selection of phylogenetic informative regions from multiple sequence alignments. BMC Evol. Biol. 10, 210. https://doi.org/10.1186/1471-2148-10-210 Crits-Christoph, A., Diamond, S., Butterfield, C.N., Thomas, B.C., Banfield, J.F., 2018. Novel soil bacteria possess diverse genes for secondary metabolite biosynthesis. Nature 558, 440–444. https://doi.org/10.1038/s41586-018-0207-y Davies‐Barnard, T., Friedlingstein, P., 2020. The Global Distribution of Biological Nitrogen Fixation in Terrestrial Natural Ecosystems. Glob. Biogeochem. Cycles 34, e2019GB006387. https://doi.org/10.1029/2019GB006387 Dos Santos, P.C., Fang, Z., Mason, S.W., Setubal, J.C., Dixon, R., 2012. Distribution of nitrogen fixation and nitrogenase-like sequences amongst microbial genomes. BMC Genomics 13, 162. https://doi.org/10.1186/1471-2164-13-162 Edgar, R.C., 2013. UPARSE: highly accurate OTU sequences from microbial amplicon reads. Nat. Methods 10, 996–998. https://doi.org/10.1038/nmeth.2604 Edgar, R.C., 2010. Search and clustering orders of magnitude faster than BLAST. Bioinformatics 26, 2460–2461. https://doi.org/10.1093/bioinformatics/btq461 Feng, X., Cheng, H., Portik, D., Li, H., 2022. Metagenome assembly of high-fidelity long reads with hifiasm-meta. Nat. Methods 19, 671–674. https://doi.org/10.1038/s41592-022-01478-3 Finn, R.D., Clements, J., Eddy, S.R., 2011. HMMER web server: interactive sequence similarity searching. Nucleic Acids Res. 39, W29–W37. https://doi.org/10.1093/nar/gkr367 Galagan, J.E., Henn, M.R., Ma, L.-J., Cuomo, C.A., Birren, B., 2005. Genomics of the fungal kingdom: Insights into eukaryotic biology. Genome Res. 15, 1620–1631. https://doi.org/10.1101/gr.3767105 Grant, J.R., Enns, E., Marinier, E., Mandal, A., Herman, E.K., Chen, C., Graham, M., Van Domselaar, G., Stothard, P., 2023. Proksee: in-depth characterization and visualization of bacterial genomes. Nucleic Acids Res. 51, W484–W492. https://doi.org/10.1093/nar/gkad326 Hunt, M., Silva, N.D., Otto, T.D., Parkhill, J., Keane, J.A., Harris, S.R., 2015. Circlator: automated circularization of genome assemblies using long sequencing reads. Genome Biol. 16, 294. https://doi.org/10.1186/s13059-015-0849-0 Huson, D.H., Beier, S., Flade, I., Górska, A., El-Hadidi, M., Mitra, S., Ruscheweyh, H.-J., Tappu, R., 2016. MEGAN Community Edition - Interactive Exploration and Analysis of Large-Scale Microbiome Sequencing Data. PLOS Comput. Biol. 12, e1004957. https://doi.org/10.1371/journal.pcbi.1004957 Jiang, F., Li, Q., Wang, S., Shen, T., Wang, H., Wang, A., Xu, D., Yuan, L., Lei, L., Chen, R., Yang, B., Deng, Y., Fan, W., 2023. Recovery of metagenome-assembled microbial genomes from a full-scale biogas plant of food waste by pacific biosciences high-fidelity sequencing. Front. Microbiol. 13, 1095497. https://doi.org/10.3389/fmicb.2022.1095497 Johnson, D.C., Unciuleac, M.-C., Dean, D.R., 2006. Controlled Expression and Functional Analysis of Iron-Sulfur Cluster Biosynthetic Components within Azotobacter vinelandii . J. Bacteriol. 188, 7551–7561. https://doi.org/10.1128/JB.00596-06 Katoh, K., Rozewicki, J., Yamada, K.D., 2019. MAFFT online service: multiple sequence alignment, interactive sequence choice and visualization. Brief. Bioinform. 20, 1160–1166. https://doi.org/10.1093/bib/bbx108 Kim, C.Y., Ma, J., Lee, I., 2022. HiFi metagenomic sequencing enables assembly of accurate and complete genomes from human gut microbiota. Nat. Commun. 13, 6367. https://doi.org/10.1038/s41467-022-34149-0 Kolmogorov, M., Bickhart, D.M., Behsaz, B., Gurevich, A., Rayko, M., Shin, S.B., Kuhn, K., Yuan, J., Polevikov, E., Smith, T.P.L., Pevzner, P.A., 2020. metaFlye: scalable long-read metagenome assembly using repeat graphs. Nat. Methods 17, 1103–1110. https://doi.org/10.1038/s41592-020-00971-x Kuroda, K., Yamamoto, K., Nakai, R., Hirakata, Y., Kubota, K., Nobu, M.K., Narihiro, T., 2022. Symbiosis between Candidatus Patescibacteria and Archaea Discovered in Wastewater-Treating Bioreactors. mBio 13, e01711-22. https://doi.org/10.1128/mbio.01711-22 Langmead, B., Salzberg, S.L., 2012. Fast gapped-read alignment with Bowtie 2. Nat. Methods 9, 357–359. https://doi.org/10.1038/nmeth.1923 Lemos, L.N., Mendes, L.W., Baldrian, P., Pylro, V.S., 2021. Genome-Resolved Metagenomics Is Essential for Unlocking the Microbial Black Box of the Soil. Trends Microbiol. 29, 279–282. https://doi.org/10.1016/j.tim.2021.01.013 Li, D., Liu, C.-M., Luo, R., Sadakane, K., Lam, T.-W., 2015. MEGAHIT: an ultra-fast single-node solution for large and complex metagenomics assembly via succinct de Bruijn graph. Bioinformatics 31, 1674–1676. https://doi.org/10.1093/bioinformatics/btv033 Li, H., 2021. New strategies to improve minimap2 alignment accuracy. Bioinformatics 37, 4572–4574. https://doi.org/10.1093/bioinformatics/btab705 López-Mondéjar, R., Brabcová, V., Štursová, M., Davidová, A., Jansa, J., Cajthaml, T., Baldrian, P., 2018. Decomposer food web in a deciduous forest shows high share of generalist microorganisms and importance of microbial biomass recycling. ISME J. 12, 1768–1778. https://doi.org/10.1038/s41396-018-0084-2 López-Mondéjar, R., Tláskal, V., Da Rocha, U.N., Baldrian, P., 2022. Global Distribution of Carbohydrate Utilization Potential in the Prokaryotic Tree of Life. mSystems 7, e00829-22. https://doi.org/10.1128/msystems.00829-22 Ma, B., Lu, C., Wang, Y., Yu, J., Zhao, K., Xue, R., Ren, H., Lv, X., Pan, R., Zhang, J., Zhu, Y., Xu, J., 2023. A genomic catalogue of soil microbiomes boosts mining of biodiversity and genetic resources. Nat. Commun. 14, 7318. https://doi.org/10.1038/s41467-023-43000-z Marcon, E., Herault, B., 2015. entropart: An R Package to Measure and Partition Diversity. J. Stat. Softw. 68, 1–26. Marx, V., 2023. Method of the year: long-read sequencing. Nat. Methods 20, 6–11. https://doi.org/10.1038/s41592-022-01730-w McMurdie, P.J., Holmes, S., 2013. phyloseq: An R package for reproducible interactive analysis and graphics of microbiome census data. PLoS ONE 8. Muñoz-Dorado, J., Marcos-Torres, F.J., García-Bravo, E., Moraleda-Muñoz, A., Pérez, J., 2016. Myxobacteria: Moving, Killing, Feeding, and Surviving Together. Front. Microbiol. 7. https://doi.org/10.3389/fmicb.2016.00781 Nan, B., Bandaria, J.N., Moghtaderi, A., Sun, I.-H., Yildiz, A., Zusman, D.R., 2013. Flagella stator homologs function as motors for myxobacterial gliding motility by moving in helical trajectories. Proc. Natl. Acad. Sci. 110. https://doi.org/10.1073/pnas.1219982110 Navarro-Muñoz, J.C., Selem-Mojica, N., Mullowney, M.W., Kautsar, S.A., Tryon, J.H., Parkinson, E.I., De Los Santos, E.L.C., Yeong, M., Cruz-Morales, P., Abubucker, S., Roeters, A., Lokhorst, W., Fernandez-Guerra, A., Cappelini, L.T.D., Goering, A.W., Thomson, R.J., Metcalf, W.W., Kelleher, N.L., Barona-Gomez, F., Medema, M.H., 2020. A computational framework to explore large-scale biosynthetic diversity. Nat. Chem. Biol. 16, 60–68. https://doi.org/10.1038/s41589-019-0400-9 Nayfach, S., Roux, S., Seshadri, R., Udwary, D., Varghese, N., Schulz, F., Wu, D., Paez-Espino, D., Chen, I.-M., Huntemann, M., Palaniappan, K., Ladau, J., Mukherjee, S., Reddy, T.B.K., Nielsen, T., Kirton, E., Faria, J.P., Edirisinghe, J.N., Henry, C.S., Jungbluth, S.P., Chivian, D., Dehal, P., Wood-Charlson, E.M., Arkin, A.P., Tringe, S.G., Visel, A., IMG/M Data Consortium, Abreu, H., Acinas, S.G., Allen, E., Allen, M.A., Alteio, L.V., Andersen, G., Anesio, A.M., Attwood, G., Avila-Magaña, V., Badis, Y., Bailey, J., Baker, B., Baldrian, P., Barton, H.A., Beck, D.A.C., Becraft, E.D., Beller, H.R., Beman, J.M., Bernier-Latmani, R., Berry, T.D., Bertagnolli, A., Bertilsson, S., Bhatnagar, J.M., Bird, J.T., Blanchard, J.L., Blumer-Schuette, S.E., Bohannan, B., Borton, M.A., Brady, A., Brawley, S.H., Brodie, J., Brown, S., Brum, J.R., Brune, A., Bryant, D.A., Buchan, A., Buckley, D.H., Buongiorno, J., Cadillo-Quiroz, H., Caffrey, S.M., Campbell, A.N., Campbell, B., Carr, S., Carroll, J., Cary, S.C., Cates, A.M., Cattolico, R.A., Cavicchioli, R., Chistoserdova, L., Coleman, M.L., Constant, P., Conway, J.M., Mac Cormack, W.P., Crowe, S., Crump, B., Currie, C., Daly, R., DeAngelis, K.M., Denef, V., Denman, S.E., Desta, A., Dionisi, H., Dodsworth, J., Dombrowski, N., Donohue, T., Dopson, M., Driscoll, T., Dunfield, P., Dupont, C.L., Dynarski, K.A., Edgcomb, V., Edwards, E.A., Elshahed, M.S., Figueroa, I., Flood, B., Fortney, N., Fortunato, C.S., Francis, C., Gachon, C.M.M., Garcia, S.L., Gazitua, M.C., Gentry, T., Gerwick, L., Gharechahi, J., Girguis, P., Gladden, J., Gradoville, M., Grasby, S.E., Gravuer, K., Grettenberger, C.L., Gruninger, R.J., Guo, J., Habteselassie, M.Y., Hallam, S.J., Hatzenpichler, R., Hausmann, B., Hazen, T.C., Hedlund, B., Henny, C., Herfort, L., Hernandez, M., Hershey, O.S., Hess, M., Hollister, E.B., Hug, L.A., Hunt, D., Jansson, J., Jarett, J., Kadnikov, V.V., Kelly, C., Kelly, R., Kelly, W., Kerfeld, C.A., Kimbrel, J., Klassen, J.L., Konstantinidis, K.T., Lee, L.L., Li, W.-J., Loder, A.J., Loy, A., Lozada, M., MacGregor, B., Magnabosco, C., Maria Da Silva, A., McKay, R.M., McMahon, K., McSweeney, C.S., Medina, M., Meredith, L., Mizzi, J., Mock, T., Momper, L., Moran, M.A., Morgan-Lang, C., Moser, D., Muyzer, G., Myrold, D., Nash, M., Nesbø, C.L., Neumann, A.P., Neumann, R.B., Noguera, D., Northen, T., Norton, J., Nowinski, B., Nüsslein, K., O’Malley, M.A., Oliveira, R.S., Maia De Oliveira, V., Onstott, T., Osvatic, J., Ouyang, Y., Pachiadaki, M., Parnell, J., Partida-Martinez, L.P., Peay, K.G., Pelletier, D., Peng, X., Pester, M., Pett-Ridge, J., Peura, S., Pjevac, P., Plominsky, A.M., Poehlein, A., Pope, P.B., Ravin, N., Redmond, M.C., Reiss, R., Rich, V., Rinke, C., Rodrigues, J.L.M., Rodriguez-Reillo, W., Rossmassler, K., Sackett, J., Salekdeh, G.H., Saleska, S., Scarborough, M., Schachtman, D., Schadt, C.W., Schrenk, M., Sczyrba, A., Sengupta, A., Setubal, J.C., Shade, A., Sharp, C., Sherman, D.H., Shubenkova, O.V., Sierra-Garcia, I.N., Simister, R., Simon, H., Sjöling, S., Slonczewski, J., Correa De Souza, R.S., Spear, J.R., Stegen, J.C., Stepanauskas, R., Stewart, F., Suen, G., Sullivan, M., Sumner, D., Swan, B.K., Swingley, W., Tarn, J., Taylor, G.T., Teeling, H., Tekere, M., Teske, A., Thomas, T., Thrash, C., Tiedje, J., Ting, C.S., Tully, B., Tyson, G., Ulloa, O., Valentine, D.L., Van Goethem, M.W., VanderGheynst, J., Verbeke, T.J., Vollmers, J., Vuillemin, A., Waldo, N.B., Walsh, D.A., Weimer, B.C., Whitman, T., Van Der Wielen, P., Wilkins, M., Williams, T.J., Woodcroft, B., Woolet, J., Wrighton, K., Ye, J., Young, E.B., Youssef, N.H., Yu, F.B., Zemskaya, T.I., Ziels, R., Woyke, T., Mouncey, N.J., Ivanova, N.N., Kyrpides, N.C., Eloe-Fadrosh, E.A., 2021. A genomic catalog of Earth’s microbiomes. Nat. Biotechnol. 39, 499–509. https://doi.org/10.1038/s41587-020-0718-6 Nguyen, L.-T., Schmidt, H.A., Von Haeseler, A., Minh, B.Q., 2015. IQ-TREE: A Fast and Effective Stochastic Algorithm for Estimating Maximum-Likelihood Phylogenies. Mol. Biol. Evol. 32, 268–274. https://doi.org/10.1093/molbev/msu300 Olm, M.R., Brown, C.T., Brooks, B., Banfield, J.F., 2017. dRep: a tool for fast and accurate genomic comparisons that enables improved genome recovery from metagenomes through de-replication. ISME J. 11, 2864–2868. https://doi.org/10.1038/ismej.2017.126 Orakov, A., Fullam, A., Coelho, L.P., Khedkar, S., Szklarczyk, D., Mende, D.R., Schmidt, T.S.B., Bork, P., 2021. GUNC: detection of chimerism and contamination in prokaryotic genomes. Genome Biol. 22, 178. https://doi.org/10.1186/s13059-021-02393-0 Parks, D.H., Imelfort, M., Skennerton, C.T., Hugenholtz, P., Tyson, G.W., 2015. CheckM: assessing the quality of microbial genomes recovered from isolates, single cells, and metagenomes. Genome Res. 25, 1043–1055. https://doi.org/10.1101/gr.186072.114 Parks, D. H., Rinke, C., Chuvochina, M., Chaumeil, P.A., Woodcroft, B.J., Evans, P.N., et al. (2017). Recovery of nearly 8,000 metagenome-assembled genomes substantially expands the tree of life. Nature Microbiol., 2(11), 1533-1542. https://doi.org/10.1038/s41564-017-0012-7 Paul, E.A., Frey, S.D. (Eds.), 2024. Soil microbiology, ecology, and biochemistry, Fifth edition. ed. Elsevier, Amsterdam, Netherlands. Pebesma, E., 2018. Simple Features for R: Standardized Support for Spatial Vector Data. R J. 10, 439. https://doi.org/10.32614/RJ-2018-009 Quast, C., Pruesse, E., Yilmaz, P., Gerken, J., Schweer, T., Yarza, P., Peplies, J., Glöckner, F.O., 2012. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res. 41, D590–D596. https://doi.org/10.1093/nar/gks1219 Rho, M., Tang, H., Ye, Y., 2010. FragGeneScan: predicting genes in short and error-prone reads. Nucleic Acids Res. 38, e191–e191. https://doi.org/10.1093/nar/gkq747 Rinne‐Garmston, K.T., Peltoniemi, K., Chen, J., Peltoniemi, M., Fritze, H., Mäkipää, R., 2019. Carbon flux from decomposing wood and its dependency on temperature, wood N 2 fixation rate, moisture and fungal composition in a Norway spruce forest. Glob. Change Biol. 25, 1852–1867. https://doi.org/10.1111/gcb.14594 Sagova-Mareckova, M., Cermak, L., Novotna, J., Plhackova, K., Forstova, J., Kopecky, J., 2008. Innovative Methods for Soil DNA Purification Tested in Soils with Widely Differing Characteristics. Appl. Environ. Microbiol. 74, 2902–2907. https://doi.org/10.1128/AEM.02161-07 Saraiva, J.P., Bartholomäus, A., Toscan, R.B., Baldrian, P., Nunes da Rocha, U., 2023. Recovery of 197 eukaryotic bins reveals major challenges for eukaryote genome reconstruction from terrestrial metagenomes. Mol. Ecol. Resour. 23, 1066–1076. https://doi.org/10.1111/1755-0998.13776 Seibold, S., Rammer, W., Hothorn, T., Seidl, R., Ulyshen, M.D., Lorz, J., Cadotte, M.W., Lindenmayer, D.B., Adhikari, Y.P., Aragón, R., Bae, S., Baldrian, P., Barimani Varandi, H., Barlow, J., Bässler, C., Beauchêne, J., Berenguer, E., Bergamin, R.S., Birkemoe, T., Boros, G., Brandl, R., Brustel, H., Burton, P.J., Cakpo-Tossou, Y.T., Castro, J., Cateau, E., Cobb, T.P., Farwig, N., Fernández, R.D., Firn, J., Gan, K.S., González, G., Gossner, M.M., Habel, J.C., Hébert, C., Heibl, C., Heikkala, O., Hemp, A., Hemp, C., Hjältén, J., Hotes, S., Kouki, J., Lachat, T., Liu, J., Liu, Y., Luo, Y.-H., Macandog, D.M., Martina, P.E., Mukul, S.A., Nachin, B., Nisbet, K., O’Halloran, J., Oxbrough, A., Pandey, J.N., Pavlíček, T., Pawson, S.M., Rakotondranary, J.S., Ramanamanjato, J.-B., Rossi, L., Schmidl, J., Schulze, M., Seaton, S., Stone, M.J., Stork, N.E., Suran, B., Sverdrup-Thygeson, A., Thorn, S., Thyagarajan, G., Wardlaw, T.J., Weisser, W.W., Yoon, S., Zhang, N., Müller, J., 2021. The contribution of insects to global forest deadwood decomposition. Nature 597, 77–81. https://doi.org/10.1038/s41586-021-03740-8 Sereika, M., Kirkegaard, R.H., Karst, S.M., Michaelsen, T.Y., Sørensen, E.A., Wollenberg, R.D., Albertsen, M., 2022. Oxford Nanopore R10.4 long-read sequencing enables the generation of near-finished bacterial genomes from pure cultures and metagenomes without short-read or reference polishing. Nat. Methods 19, 823–826. https://doi.org/10.1038/s41592-022-01539-7 Shaffer, M., Borton, M.A., McGivern, B.B., Zayed, A.A., La Rosa, S.L., Solden, L.M., Liu, P., Narrowe, A.B., Rodríguez-Ramos, J., Bolduc, B., Gazitúa, M.C., Daly, R.A., Smith, G.J., Vik, D.R., Pope, P.B., Sullivan, M.B., Roux, S., Wrighton, K.C., 2020. DRAM for distilling microbial metabolism to automate the curation of microbiome function. Nucleic Acids Res. 48, 8883–8900. https://doi.org/10.1093/nar/gkaa621 Sharrar, A.M., Crits-Christoph, A., Méheust, R., Diamond, S., Starr, E.P., Banfield, J.F., 2020. Bacterial Secondary Metabolite Biosynthetic Potential in Soil Varies with Phylum, Depth, and Vegetation Type. mBio 11, e00416-20. https://doi.org/10.1128/mBio.00416-20 Shen, W., Le, S., Li, Y., & Hu, F., 2016. SeqKit: a cross-platform and ultrafast toolkit for FASTA/Q file manipulation. PloS one, 11(10), e0163962. https://doi.org/10.1371/journal.pone.0163962 Terlouw, B.R., Blin, K., Navarro-Muñoz, J.C., Avalon, N.E., Chevrette, M.G., Egbert, S., Lee, S., Meijer, D., Recchia, M.J.J., Reitz, Z.L., van Santen, J.A., Selem-Mojica, N., Tørring, T., Zaroubi, L., Alanjary, M., Aleti, G., Aguilar, C., Al-Salihi, S.A.A., Augustijn, H.E., Avelar-Rivas, J.A., Avitia-Domínguez, L.A., Barona-Gómez, F., Bernaldo-Agüero, J., Bielinski, V.A., Biermann, F., Booth, T.J., Carrion Bravo, V.J., Castelo-Branco, R., Chagas, F.O., Cruz-Morales, P., Du, C., Duncan, K.R., Gavriilidou, A., Gayrard, D., Gutiérrez-García, K., Haslinger, K., Helfrich, E.J.N., van der Hooft, J.J.J., Jati, A.P., Kalkreuter, E., Kalyvas, N., Kang, K.B., Kautsar, S., Kim, W., Kunjapur, A.M., Li, Y.-X., Lin, G.-M., Loureiro, C., Louwen, J.J.R., Louwen, N.L.L., Lund, G., Parra, J., Philmus, B., Pourmohsenin, B., Pronk, L.J.U., Rego, A., Rex, D.A.B., Robinson, S., Rosas-Becerra, L.R., Roxborough, E.T., Schorn, M.A., Scobie, D.J., Singh, K.S., Sokolova, N., Tang, X., Udwary, D., Vigneshwari, A., Vind, K., Vromans, S.P.J.M., Waschulin, V., Williams, S.E., Winter, J.M., Witte, T.E., Xie, H., Yang, D., Yu, J., Zdouc, M., Zhong, Z., Collemare, J., Linington, R.G., Weber, T., Medema, M.H., 2023. MIBiG 3.0: a community-driven effort to annotate experimentally validated biosynthetic gene clusters. Nucleic Acids Res. 51, D603–D610. https://doi.org/10.1093/nar/gkac1049 Thiery, S., Kaimer, C., 2020. The Predation Strategy of Myxococcus xanthus. Front. Microbiol. 11, 2. https://doi.org/10.3389/fmicb.2020.00002 Tláskal, V., Baldrian, P., 2021. Deadwood-Inhabiting Bacteria Show Adaptations to Changing Carbon and Nitrogen Availability During Decomposition. Front. Microbiol. 12, 685303. https://doi.org/10.3389/fmicb.2021.685303 Tláskal, V., Brabcová, V., Větrovský, T., Jomura, M., López-Mondéjar, R., Oliveira Monteiro, L.M., Saraiva, J.P., Human, Z.R., Cajthaml, T., Nunes Da Rocha, U., Baldrian, P., 2021a. Complementary Roles of Wood-Inhabiting Fungi and Bacteria Facilitate Deadwood Decomposition. mSystems 6, e01078-20. https://doi.org/10.1128/mSystems.01078-20 Tláskal, V., Brabcová, V., Větrovský, T., López-Mondéjar, R., Monteiro, L.M.O., Saraiva, J.P., Da Rocha, U.N., Baldrian, P., 2021b. Metagenomes, metatranscriptomes and microbiomes of naturally decomposing deadwood. Sci. Data 8, 198. https://doi.org/10.1038/s41597-021-00987-8 Tláskal, V., Zrůstová, P., Vrška, T., Baldrian, P., 2017. Bacteria associated with decomposing dead wood in a natural temperate forest. FEMS Microbiol. Ecol. 93. https://doi.org/10.1093/femsec/fix157 Tokumoto, U., 2004. Interchangeability and Distinct Properties of Bacterial Fe-S Cluster Assembly Systems: Functional Replacement of the isc and suf Operons in Escherichia coli with the nifSU-Like Operon from Helicobacter pylori. J. Biochem. (Tokyo) 136, 199–209. https://doi.org/10.1093/jb/mvh104 Tria, F.D.K., Landan, G., Dagan, T., 2017. Phylogenetic rooting using minimal ancestor deviation. Nat. Ecol. Evol. 1, 0193. https://doi.org/10.1038/s41559-017-0193 Uliano-Silva, M., Ferreira, J.G.R.N., Krasheninnikova, K., Darwin Tree of Life Consortium, Blaxter, M., Mieszkowska, N., Hall, N., Holland, P., Durbin, R., Richards, T., Kersey, P., Hollingsworth, P., Wilson, W., Twyford, A., Gaya, E., Lawniczak, M., Lewis, O., Broad, G., Martin, F., Hart, M., Barnes, I., Formenti, G., Abueg, L., Torrance, J., Myers, E.W., Durbin, R., Blaxter, M., McCarthy, S.A., 2023. MitoHiFi: a python pipeline for mitochondrial genome assembly from PacBio high fidelity reads. BMC Bioinformatics 24, 288. https://doi.org/10.1186/s12859-023-05385-y Uritskiy, G.V., DiRuggiero, J., Taylor, J., 2018. MetaWRAP—a flexible pipeline for genome-resolved metagenomic data analysis. Microbiome 6, 158. https://doi.org/10.1186/s40168-018-0541-1 Větrovský, T., Baldrian, P., Morais, D., 2018. SEED 2: a user-friendly platform for amplicon high-throughput sequencing data analyses. Bioinformatics 34, 2292–2294. https://doi.org/10.1093/bioinformatics/bty071 Wang, Y., Gallagher, L.A., Andrade, P.A., Liu, A., Humphreys, I.R., Turkarslan, S., Cutler, K.J., Arrieta-Ortiz, M.L., Li, Y., Radey, M.C., McLean, J.S., Cong, Q., Baker, D., Baliga, N.S., Peterson, S.B., Mougous, J.D., 2023. Genetic manipulation of Patescibacteria provides mechanistic insights into microbial dark matter and the epibiotic lifestyle. Cell 186, 4803-4817.e13. https://doi.org/10.1016/j.cell.2023.08.017 Waschulin, V., Borsetto, C., James, R., Newsham, K.K., Donadio, S., Corre, C., Wellington, E., 2022. Biosynthetic potential of uncultured Antarctic soil bacteria revealed through long-read metagenomic sequencing. ISME J. 16, 101–111. https://doi.org/10.1038/s41396-021-01052-3 West, P.T., Probst, A.J., Grigoriev, I.V., Thomas, B.C., Banfield, J.F., 2018. Genome-reconstruction for eukaryotes from complex natural microbial communities. Genome Res. 28, 569–580. https://doi.org/10.1101/gr.228429.117 Wick, R.R., Judd, L.M., Gorrie, C.L., Holt, K.E., 2017. Unicycler: Resolving bacterial genome assemblies from short and long sequencing reads. PLOS Comput. Biol. 13, e1005595. https://doi.org/10.1371/journal.pcbi.1005595 Wood, D.E., Lu, J., Langmead, B., 2019. Improved metagenomic analysis with Kraken 2. Genome Biol. 20, 257. https://doi.org/10.1186/s13059-019-1891-0 Woodcroft, B.J., Singleton, C.M., Boyd, J.A., Evans, P.N., Emerson, J.B., Zayed, A.A.F., Hoelzle, R.D., Lamberton, T.O., McCalley, C.K., Hodgkins, S.B., Wilson, R.M., Purvine, S.O., Nicora, C.D., Li, C., Frolking, S., Chanton, J.P., Crill, P.M., Saleska, S.R., Rich, V.I., Tyson, G.W., 2018. Genome-centric view of carbon processing in thawing permafrost. Nature 560, 49–54. https://doi.org/10.1038/s41586-018-0338-1 Zehr, J.P., Capone, D.G., 2020. Changing perspectives in marine nitrogen fixation. Science 368, eaay9514. https://doi.org/10.1126/science.aay9514 Zhang, H., Yohe, T., Huang, L., Entwistle, S., Wu, P., Yang, Z., Busk, P.K., Xu, Y., Yin, Y., 2018. dbCAN2: a meta server for automated carbohydrate-active enzyme annotation. Nucleic Acids Res. 46, W95–W101. Additional Declarations No competing interests reported. Supplementary Files Supplementaryfiguresdeadwoodpaper20240328.pdf Supplementarytablesdeadwoodpaper20240328.xlsx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4181686","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":286067015,"identity":"e1243a6b-487d-49bd-bdb2-f84528272b9e","order_by":0,"name":"Etienne Richy","email":"","orcid":"","institution":"Institute of Microbiology of the Czech Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Etienne","middleName":"","lastName":"Richy","suffix":""},{"id":286067016,"identity":"8fe2282a-aa29-4cb7-9519-80c6da6f461c","order_by":1,"name":"Priscila Thiago Dobbler","email":"","orcid":"","institution":"Institute of Microbiology of the Czech Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Priscila","middleName":"Thiago","lastName":"Dobbler","suffix":""},{"id":286067017,"identity":"2bdb8cd7-1f88-4bfa-84cc-2a60178344ef","order_by":2,"name":"Vojtěch Tláskal","email":"","orcid":"","institution":"Institute of Microbiology of the Czech Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Vojtěch","middleName":"","lastName":"Tláskal","suffix":""},{"id":286067018,"identity":"822e7053-30f2-470c-98be-f03ab56531ea","order_by":3,"name":"Rubén López-Mondéjar","email":"","orcid":"","institution":"Institute of Microbiology of the Czech Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Rubén","middleName":"","lastName":"López-Mondéjar","suffix":""},{"id":286067019,"identity":"351e4029-73d8-405a-b260-9ac79175e189","order_by":4,"name":"Petr Baldrian","email":"","orcid":"","institution":"Institute of Microbiology of the Czech Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Petr","middleName":"","lastName":"Baldrian","suffix":""},{"id":286067020,"identity":"ab29e7a6-d9a2-42e1-8a7e-07f7376ac030","order_by":5,"name":"Martina Kyselková","email":"data:image/png;base64,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","orcid":"","institution":"Institute of Microbiology of the Czech Academy of Sciences","correspondingAuthor":true,"prefix":"","firstName":"Martina","middleName":"","lastName":"Kyselková","suffix":""}],"badges":[],"createdAt":"2024-03-28 10:57:18","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4181686/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4181686/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":54097813,"identity":"3f6e94c7-9cce-455e-a390-54252d844d13","added_by":"auto","created_at":"2024-04-04 14:48:20","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":566027,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSampling strategy and metagenomic workflow.\u003cbr\u003e\n \u003c/strong\u003e\u003cem\u003eDeadwood samples (n = 4) were collected from the Žofínský Prales National Nature Reserve, Czech Republic. For each sample, Fagus sylvatica trunks were drilled vertically from the middle of the top surface at five equidistant locations, and sawdust from the inside of the trunks was collected and pooled. The four samples were sequenced using Pacbio HiFi and Illumina HiSeq. Sample-by-sample assembly and co-assembly was performed using short and long read, as well as hybrid assembly. Assemblies were binned independently and bins were pooled, with the exception of the hybrid assembly, which gave unsatisfactory results. After quality control (completeness and contamination, dereplication and mis-binning detection), the phylogeny and metabolic capacities of the metagenomes-assembled genomes were analyzed. Transcripts were mapped to MAGs to validate the functionality of specific metabolism. The figure illustrating the metagenomic workflow was created with Biorender.com.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-4181686/v1/829c6e29eebec7672d236e52.png"},{"id":54097814,"identity":"d30ff07e-0208-49dd-bf2f-87a5fac5d276","added_by":"auto","created_at":"2024-04-04 14:48:20","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":503760,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eIntegrating \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003ede novo \u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003egenome assembly and metabarcoding.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e\u003cstrong\u003eA. \u003c/strong\u003e\u003c/em\u003e\u003cem\u003eRelative abundance of sequences (%), phylum richness (H0) and alpha diversity (H1) index estimated from 16S rRNA amplicon metabarcoding, and the number of recovered MAGs per bacterial phyla. \u003c/em\u003e\u003cem\u003e\u003cstrong\u003eB.\u003c/strong\u003e\u003c/em\u003e\u003cem\u003eMaximum likelihood phylogenomic tree, classifying MAGs based on their position in the reference tree, relative evolutionary divergence, and average nucleotide identity (ANI) compared to reference genomes. The tree was generated using GTDB-tk based on the concatenated phylogeny of 120 bacterial single-copy marker genes and includes 69 unique bacterial genomes (MAGs) with completeness \u0026gt; 50% and contamination \u0026lt; 10%. \u003c/em\u003e\u003cem\u003e\u003cstrong\u003eC. \u003c/strong\u003e\u003c/em\u003e\u003cem\u003eBarplot displaying the averaged OTU relative abundance across all four samples. Only OTUs with an average relative abundance \u0026gt; 0.1% (top 172) are shown and ranked in descending order. The bar color corresponds to the phylum when a corresponding MAG was assembled. \u003c/em\u003e\u003cem\u003e\u003cstrong\u003eD.\u003c/strong\u003e\u003c/em\u003e\u003cem\u003eCorrelation between MAG relative abundance and sequence relative abundance of corresponding OTUs.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-4181686/v1/20d58feea9198b0f022b0dc4.png"},{"id":54097821,"identity":"607c8b4c-42d6-43b2-b913-13a8568d4632","added_by":"auto","created_at":"2024-04-04 14:48:20","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":702745,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFunctional profiles of the MAGs assembled from deadwood.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eHeatmap representing the metabolic capacities involved in the decomposition of the easily decomposable carbohydrates (arabinogalactanases, xylanases/xyloglucanases, mannanases, cellobiases and xylobiases) and complex plant cell wall polymers (endoglucanases, cellobiohydrolases, exoglucanases), microbial biomass (peptidoglycansases, betaglucanases, chitinases), and reserve compounds (alphaglucanases), nitrogen fixation, dissimilatory nitrate reduction (nitrite to ammonia only) and assimilatory nitrate reduction (nitrite to ammonia only), and the number of biosynthetic gene clusters found in the 69 MAGs. The top row represents the phylum to which the MAG was affilicated. The number of CAZymes were log transformed +1 to improve the visualisation, and the total counts of CAZymes were summed on the right last column per activity or by class of enzyme. The bottom row represents the average relative abundance of the MAG. The shape of the bacteria was drawn based on the closest characterised clades of each MAG lineages.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-4181686/v1/e41addcaa7b40f665bae06db.png"},{"id":54097819,"identity":"24d3ec4f-e341-41a7-ad64-ba1d617c6cae","added_by":"auto","created_at":"2024-04-04 14:48:20","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":583747,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCharacterization of the genomic regions associated with N\u003c/strong\u003e\u003csub\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/sub\u003e\u003cstrong\u003e fixation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e\u003cstrong\u003eA. \u003c/strong\u003e\u003c/em\u003e\u003cem\u003eGene organization of the five genomic regions associated with N\u003c/em\u003e\u003csub\u003e\u003cem\u003e2\u003c/em\u003e\u003c/sub\u003e\u003cem\u003e fixation assembled from PacBio HiFi assembly. Clustering is based on the phylogenetic relationships of the corresponding nifH sequences. Only regions associated with N\u003c/em\u003e\u003csub\u003e\u003cem\u003e2\u003c/em\u003e\u003c/sub\u003e\u003cem\u003e fixation were plotted. The genes are coloured by operons and systems: in cyan the nif genes, in blue the fix genes, in red the Isc system genes and in green the Suf system genes. \u003c/em\u003e\u003cem\u003e\u003cstrong\u003eB. \u003c/strong\u003e\u003c/em\u003e\u003cem\u003eNumbers of transcripts mapping to the nitrogenase genes (nifHDK) found in the five contigs (from A), belonging either to Alphaproteobacteria (green) or Gammaproteobacteria (yellow). The contig s509.ctg000513l corresponds to the contig found in the MAG 36_Steroidobacteraceae. \u003c/em\u003e\u003cem\u003e\u003cstrong\u003eC.\u003c/strong\u003e\u003c/em\u003e\u003cem\u003e Circular plot illustrating genomic islands of the MAG 36_Steroidobacteraceae computed with Islandviewer4. The exact location of the nitrogen fixation contig s509.ctg000513l is between 5118742 and 5335241 bp, where no genomic island was detected. \u003c/em\u003e\u003cem\u003e\u003cstrong\u003eD.\u003c/strong\u003e\u003c/em\u003e\u003cem\u003e Phylogenetic analysis of the nifH gene found in the MAG 36_Steroidobacteraceae. Neighbor-joining trees were generated using the maximum likelihood algorithm with 1,000 bootstrap iterations and rooted using minimal ancestor deviation. First, the nifH sequences from the InterPro database (accessed on August 1st, 2023) clustered at 97% of similarity were included, followed by sub-selection of the closest related taxa.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-4181686/v1/20fcbe3441583b6a9071ba31.png"},{"id":54097816,"identity":"ef8f4b1e-9f08-449b-96da-9f2c2690718b","added_by":"auto","created_at":"2024-04-04 14:48:20","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":327825,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eBiosynthetic diversity in MAGs.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e\u003cstrong\u003eA\u003c/strong\u003e\u003c/em\u003e\u003cem\u003e. Barplot summarizing the number of BGCs identified in MAGs derived from long read co-assembly. BGCs are categorized by type and ranked in descending order. Colors indicate the number of BGCs per phylum. TPS, terpenes; RiPP, ribosomally synthesised and post-translationally modified peptides; PKS, polyketide synthase product; NRPS, non-ribosomal peptide synthetase product;\u003c/em\u003e \u003cem\u003eLAP, linear azol(in)e-containing peptides; NAPAA, non-alpha poly-amino acids like e-polylysin; CDPS, tRNA-dependent cyclodipeptide synthase product. \u003c/em\u003e\u003cem\u003e\u003cstrong\u003eB\u003c/strong\u003e\u003c/em\u003e\u003cem\u003e. Similarity network of 279 BGCs found in 69 MAGs. BGCs belonging to the same gene cluster family are linked. Gene cluster families with more than 2 members are labeled with the corresponding product type and taxonomic classification of the MAGs in which they occurred (in parenthesis, all BGCs within one family always originated from the same genus). \u003c/em\u003e\u003cem\u003e\u003cstrong\u003eC\u003c/strong\u003e\u003c/em\u003e\u003cem\u003e. Organization of BGCs encoding thioamitides. Seven thioamitide BGCs found in 5 MAGs (each belonging to a unique gene cluster family) are displayed together with thioviridamide BGC from Streptomyces olivovoridis as a reference. The core biosynthetic genes ycaO and tfuA are labeled.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-4181686/v1/60dfa205e1e51a46660a6aeb.png"},{"id":55923073,"identity":"7bbf3d34-5983-4e59-8022-ce74b06eca07","added_by":"auto","created_at":"2024-05-06 10:43:17","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1589295,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4181686/v1/d24cb746-f5aa-4e62-8128-963a5f0656b7.pdf"},{"id":54098195,"identity":"ef140f96-ed25-40fc-aeb8-5cd1c5dabe49","added_by":"auto","created_at":"2024-04-04 14:56:20","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":895580,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementaryfiguresdeadwoodpaper20240328.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4181686/v1/43dbb1197f559ed099db65d8.pdf"},{"id":54097818,"identity":"6745d51e-8690-4490-b51f-02595d4927ea","added_by":"auto","created_at":"2024-04-04 14:48:20","extension":"xlsx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":608482,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementarytablesdeadwoodpaper20240328.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-4181686/v1/beec971c9de6e7f0d979b749.xlsx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Pacbio HiFi sequencing sheds light on key bacteria contributing to deadwood decomposition processes","fulltext":[{"header":"Introduction","content":"\u003cp\u003eFree-living microorganisms play a crucial role in global nutrient cycling, and their diverse metabolic abilities hold potential solutions for environmental and health issues. However, understanding microbial functional features in natural habitats is challenging due to the complexity of environmental microbiomes and the limitations of cultivation methods. Shotgun metagenomics offers an alternative to culturing, providing insights into taxonomic composition and functional potential of microbial species through the construction of metagenome-assembled genomes (MAGs) (Lemos et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Unfortunately, due to the limitations of assembly performance, MAGs from second-generation (short-read) sequencing suffer from incompleteness, high level of fragmentation and contamination. Consequently, functional inferences may be misleading, the assembly of complete operons or large genomic islands is mostly impossible and identification is unclear since discriminative genes such as the 16S rRNA are largely missing (Parks et al., \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2017\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThird-generation (long-read) sequencing allows much more reliable assembly and thus holds the promise for recovering high-quality MAGs or complete/circular prokaryotic genomes from complex microbiome samples (Bickhart et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Sereika et al., \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), enabling an accurate assessment of their ecological roles. The two main long-read platforms include the cost-affordable Oxford Nanopore, albeit with inferior accuracy (Brown et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Sereika et al., \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), and the PacBio platform, offering highly accurate consensus sequencing (HiFi) reads without the need for further polishing (Jiang et al., \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Kim et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn this study, we demonstrate the added value of PacBio HiFi sequencing over short-read sequencing for characterizing dominant microbiome members in decomposing deadwood. Deadwood, characterized by high carbon and low nitrogen concentrations, offers a variety of niches for microorganisms and represents a hotspot of biodiversity (Seibold et al., \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Although deadwood degradation contributes to release of CO\u003csub\u003e2\u003c/sub\u003e into the atmosphere, its final degradation products also enrich soils with nutrients and recalcitrant organic matter (OM), assisting nutrient cycling and carbon storage in forest ecosystems (Baldrian et al., \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Here, we used deadwood samples from a natural mixed temperate forest in Central Europe, which were previously analyzed through bacterial genomics, metagenomics and metatranscriptomics using second generation high-throughput sequencing (Tl\u0026aacute;skal et al., \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2021a\u003c/span\u003e, \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2021b\u003c/span\u003e, \u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). These studies identified fungi and bacteria as the main contributors to deadwood microbiomes. Fungi are efficient lignin degraders (Boer et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2005\u003c/span\u003e) and play the primary role in deadwood decomposition, while some bacteria contribute to decomposition processes by degrading cellulose, hemicellulose and fungal biomass. Importantly, bacteria have an essential role in nitrogen cycling in the deadwood, and the unique ability of some taxa to fix atmospheric nitrogen (N\u003csub\u003e2\u003c/sub\u003e) compensates the high C:N ratio typical for deadwood (Rinne-Garmston et al., \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eDespite recent progress in understanding bacterial roles in deadwood, the information gained from genomes recovered from second-generation high-throughput sequencing remained incomplete and fragmented (Tl\u0026aacute;skal et al., \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2021a\u003c/span\u003e, \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003eb\u003c/span\u003e). Complete genomes could be recovered from bacterial isolates from deadwood, but culture bias limited this approach to cultivable phyla (Tl\u0026aacute;skal and Baldrian, \u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Here, we show that PacBio HiFi sequencing of long DNA molecules overcomes these limitations, providing higher numbers of MAGs than the assembly of short reads, the MAGs being moreover more complete, less contaminated and less fragmented. Assembly of DNA contigs containing complete operons with crucial functions allow better evaluation of the functional potential of deadwood bacteria. Ultimately, long read metagenomics in combination with metatranscriptomics identified new key actors in wood decomposition that remained \u0026ldquo;hidden\u0026rdquo; for the more traditional approaches.\u003c/p\u003e"},{"header":"Material and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eSample description\u003c/h2\u003e \u003cp\u003eDeadwood samples were obtained from the core zone of the Žof\u0026iacute;nsk\u0026yacute; prales National Nature Reserve, Czech Republic (48\u0026deg;39\u0026prime;57\u0026Prime;N, 14\u0026deg;42\u0026prime;24\u0026Prime;E). The core zone of the Žof\u0026iacute;nsk\u0026yacute; prales is an unmanaged forest without any human interventions since 1838. The site and deadwood sampling were previously described in detail (Tl\u0026aacute;skal et al., \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2021a\u003c/span\u003e). A map generated with the r package \u003cem\u003esf\u003c/em\u003e (Pebesma, \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) representing the location of the sampling site is provided in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. For this study, four samples from \u003cem\u003eFagus sylvatica\u003c/em\u003e trunks representing class 2 (4\u0026ndash;7 years of decomposition; sample 6 \u0026ndash; BioSample accession SAMN13925154 and sample 7 \u0026ndash; SAMN13925155) and class 4 (20\u0026ndash;41 years of decomposition; sample 57 \u0026ndash; SAMN13925167 and sample 84 \u0026ndash; SAMN13925168) were chosen. The samples were stored at -80\u0026deg;C prior to DNA isolation. Short-read sequencing using Illumina HiSeq platform was previously performed on these samples (Tl\u0026aacute;skal et al., \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2021a\u003c/span\u003e, \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003eb\u003c/span\u003e) and the short-read data were used in this study.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003eDNA isolation\u003c/h2\u003e \u003cp\u003eHigh molecular weight DNA was isolated from the four deadwood samples using a phenol/chloroform/isoamyl alcohol extraction method according to Sagova-Mareckova et al. (\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e2008\u003c/span\u003e) with several modifications; notably we replaced the bead-beating step by vortexing in order to prevent excessive DNA shearing. Samples were first thoroughly ground in a sterile mortar with liquid nitrogen using a rough pestle. Sample aliquots (~\u0026thinsp;250 mg) were added to 2 mL tubes with a screw lid and mixed with 600 \u0026micro;L extraction buffer (50 mM Na-phosphate buffer pH 8, 50 mM NaCl, 500 mM Tris-HCl pH 8, and 5% sodium dodecyl sulfate) and 300 \u0026micro;L phenol (pH 8)/chloroform/isoamyl alcohol (25:24:1). The tubes were vortexed for 1 min at 4\u0026deg;C, using the Vortex-Genie (Scientific Industries, Bohemia, NY) set at the maximum speed. The homogenized samples were centrifuged at 10,000 g for 3 min and supernatant was transferred to a clean tube. One volume of phenol/chloroform/isoamyl alcohol (25:24:1) was added and mixed with the supernatant by gentle tube inverting for 1 min. After centrifugation at 6,000 g for 5 min, the supernatant was transferred to a clean tube, mixed (gentle tube inverting for 1 min) with 1 volume of chloroform/isoamyl alcohol (24:1), and centrifuged (6,000 g, 5 min). The supernatant was transferred to a clean tube and mixed with NaCl (to a final concentration of 1.5 M) and CTAB (to a final concentration of 1%), and incubated at 65\u0026deg;C for 35 min. The incubated solution was cooled at 4\u0026deg;C for 5 min and then mixed with an equal volume of chloroform/isoamyl alcohol (24:1), and centrifuged at 3,400 g for 20 min. The supernatant was mixed with 0.6 volume of isopropanol and 0.1 volume of 3M sodium acetate in a clean tube and incubated at 4\u0026deg;C for 45 min to precipitate the DNA. After centrifugation at 10,000 g for 20 min, the supernatant was removed and the DNA pellet was washed with 200 \u0026micro;L cold 70% ethanol, air-dried and resuspended in 30 \u0026micro;L 10 mM Tris buffer pH 8. Presence of high molecular weight DNA was verified with DNA electrophoresis in 0.8% agarose gel and DNA concentration was measured with Qubit 2.0 Fluorometer (Invitrogen, Carlsbad, CA). Three to six DNA aliquots per sample were pooled to obtain the minimal required DNA quantity of 5 \u0026micro;g.\u003c/p\u003e \u003cp\u003e \u003cem\u003eSequencing and primary analyses\u003c/em\u003e of the metagenome\u003c/p\u003e \u003cp\u003eThe library preparation and PacBio HiFi sequencing on the Sequel II Instrument were performed at Brigham Young University Sequencing Centre, Utah. Samples 6 and 7 were pooled equimolarly during library preparation and sequenced together on one 8M SMRT Cell using 30-h movie. Samples 57 and 84 were sequenced separately on one 8M SMRT Cell using 30-h movie. PacBio\u0026rsquo;s movie subreads files were processed with CCS (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/PacificBiosciences/pbbioconda\u003c/span\u003e\u003cspan address=\"https://github.com/PacificBiosciences/pbbioconda\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e)\u003c/span\u003e for consensus Hi-Fi reads of \u0026gt;\u0026thinsp;99% accuracy. After removing reads of less than 1,000 bp, a total of 16.1\u0026times;10\u003csup\u003e9\u003c/sup\u003e bp were obtained for the 4 samples (\u003cb\u003eTable \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e\u003c/b\u003e). Next, taxonomic profiling of the raw PacBio HiFi reads was performed using DIAMOND\u0026thinsp;+\u0026thinsp;MEGAN-LR workflow, where reads are aligned with DIAMOND v2.1.8 (Buchfink et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) blastx against the NCBI nr database (version 09/2022) using the following parameter \u0026lsquo;--range-culling --top 5 -F 5000\u0026rsquo; for long-read mode, followed by \u0026lsquo;meganization\u0026rsquo; where LCA (last common ancestor) of each read is assigned. For this, the MEGAN6 Community Edition (Huson et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2016\u003c/span\u003e) \u0026lsquo;daa2rma\u0026rsquo; function was used with the following parameter \u0026lsquo;--longReads --lcaAlgorithm longReads -ram readCount\u0026rsquo;.\u003c/p\u003e \u003cp\u003eThe metagenome and metatranscriptome of these samples previously sequenced on an Illumina HiSeq 2500 (2 \u0026times; 250 bases), were downloaded from Tl\u0026aacute;skal et al. (\u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2021b\u003c/span\u003e). The quality of the reads was checked using Trimmomatic (v0.36) (Bolger et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2014\u003c/span\u003e) and FASTX-Toolkit (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://hannonlab.cshl.edu/fastx_toolkit/\u003c/span\u003e\u003cspan address=\"http://hannonlab.cshl.edu/fastx_toolkit/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), removing the adaptor contamination, the low quality reads (\u0026lt;\u0026thinsp;30), the reads\u0026thinsp;\u0026lt;\u0026thinsp;50bp and trimming the low-quality ends of reads. Only the metatranscriptomes of samples 6 and 7 were available, containing 13\u0026times;10\u003csup\u003e6\u003c/sup\u003e and 71\u0026times;10\u003csup\u003e6\u003c/sup\u003e sequences, respectively (\u003cb\u003eTable \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e\u003c/b\u003e). The Illumina HiSeq sequencing of the metagenomes generated a total of 19.8 Gb. Taxonomic profiling of the raw Illumina short-read was performed using Kraken2 v2.1.2 (Wood et al., \u003cspan citationid=\"CR85\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) against the NCBI nt database using default parameters. Taxonomic profiles of raw short reads suffered from a high share (~\u0026thinsp;30%) of unassigned reads compared to raw long reads (\u003cb\u003eTable \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e\u003c/b\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003eSequence assembly and taxonomic profiling\u003c/h2\u003e \u003cp\u003eSequences generated by PacBio HiFi sequencing were first co-assembled (from the four samples) using Hifiasm-meta r058 (Feng et al., \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) with default parameters, and metaFlye v2.9.2 (Kolmogorov et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) with --pacbio-hifi and --meta setting (Kim et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Hifiasm-meta produced a total of 16,446 contigs with an N50 of 80,906bp, while metaFlye produced a total of 6,589 contigs with a N50 of 90,042bp (\u003cb\u003eTable S3\u003c/b\u003e). As a result, only Hifiasm-meta co-assemblies were further analyzed and Hifiasm-meta assembler was also used for sample-by-sample assembly. Sequences generated by Illumina HiSeq were co-assembled and sample-by-sample assembled using megahit v1.2.9 (Li et al., \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2015\u003c/span\u003e) with default settings. Hybrid assemblies were generated using Unicycler v0.5.0 (Wick et al., \u003cspan citationid=\"CR84\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) with default settings on samples 57, 6 and 7 only, as the PacBio HiFi sequencing depth of sample 84 was too low, extremely increasing runtime. The hybrid assembly of each sample was computationally more demanding than those produced with megahit and hifiasm-meta. Furthermore, as the hybrid assemblies yielded contigs shorter in length compared to PacBio HiFi and in fewer numbers than Illumina (\u003cb\u003eTable S5\u003c/b\u003e), hybrid assembly was not further investigated. Taxonomy composition of assembled contigs was based on predicted genes, by FragGeneScan v1.31 (Rho et al., \u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e2010\u003c/span\u003e), for co-assemblies and sample-by-sample assemblies. Predicted genes were aligned to NCBI nr (version 09/2022) database with DIAMOND blastp option, and taxonomy was based on best-hit (lowest e-value, highest identity and bit-score), with a minimum e-values of 10E-5.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003eMAG recovery and taxonomic identification\u003c/h2\u003e \u003cp\u003eMAGs were generated using the sample-by-sample assembly and the co-assembly of short and long reads. We employed three binning tools (MetaBAT v2.12.1, MaxBin v2.2.6, and CONCOCT v1.0.0) with default settings within metaWRAP v1.3.2 (Uritskiy et al., \u003cspan citationid=\"CR79\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). The resulting bins were then combined and refined using the Bin_refinement module of metaWRAP with the parameters \u0026lsquo;-c 50 -x 10\u0026rsquo; to obtain at least medium-quality draft MAGs. CheckM v1.0.12 (Parks et al., \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2015\u003c/span\u003e) lineage workflow in metaWRAP was used to assess the quality (completeness and contamination) of the MAGs and seqkit v2.3.0 (Shen et al., \u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e2016\u003c/span\u003e) was used for basic MAG stats. The MAGs were dereplicated using dRep v3.4.0 (Olm et al., \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Primary clusters (~\u0026thinsp;90% ANI) were formed using Mash and secondary clusters (95% ANI) were created by aligning the genomes via ANImf. Ultimately, we obtained a total of 69 species-level MAGs from PacBio HiFi long read assemblies. We used GTDB-Tk v2.1.0 (Chaumeil et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) to assign taxonomy to these MAGs based on the taxonomy R207_v2 from the Genome Taxonomy Database (GTDB). We re-estimated the completeness and the contamination of the Patescibacteria MAGs using the recently developed Checkm2 v0.1.3 (Chklovski et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), providing better estimates for this phylum. Co-assembly generated a total of 28 MAGs, with a median completeness of 65.4% and median contamination of 1.6%, including 5 of the 6 Patescibacteria and 5 of the 7 Gammaproteobacteria. Sample-by-sample assembly produced a total of 41 MAGs, with a median completeness of 71.5% and median contamination of 0.7%, including 5 of the 6 Verrucomicrobiota, 8 of the 10 Bacteroidota and the 2 Myxococcota (\u003cb\u003eTable S6\u003c/b\u003e).\u003c/p\u003e \u003cp\u003eFunctional annotation of predicted ORFs was done with DRAM v1.4.0 (Shaffer et al., \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), which relied on Pfam, KOfam, UniProt, dbCAN, and MEROPS. We estimated the relative abundances of the 69 MAGs across the four metagenome datasets using the CoverM v0.6.1 (Aroney et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) for genome abundance estimation, by assessing the coverage of mapped long reads in \u0026lsquo;genome\u0026rsquo; mode. Results expressed in relative abundance were calculated by dividing the average genome coverage by the average total coverage of all genomes, then normalized by the ratio of mapped reads to total reads. Further contamination assessment and chimerism check was done using GUNC v1.0.5 (Orakov et al., \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). We screened for the presence of ribosomal RNAs (i.e., 16S, 23S and/or 5S rRNA) in MAGs using barrnap v0.9 (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/tseemann/barrnap\u003c/span\u003e\u003cspan address=\"https://github.com/tseemann/barrnap\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eIntegrating MAGs and metabarcoding data\u003c/h2\u003e \u003cp\u003eThe prokaryotic diversity of the samples was estimated using 16S rRNA amplicon sequencing data and compared with the 16S rRNA profiles of MAGs. First, Illumina MiSeq (2 \u0026times; 250 bases) metabarcoding data for these samples were obtained from Tl\u0026aacute;skal et al., (\u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2021b\u003c/span\u003e). Subsequently, the amplicon sequencing data were processed using the SEED v2.1.05 (Větrovsk\u0026yacute; et al., \u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) pipeline. Paired-end reads were joined using fastq-join, and sequences with ambiguous bases or a mean quality score below 30 were excluded. Chimeric sequences were identified and removed using USEARCH v11.0.667 (Edgar, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2010\u003c/span\u003e), followed by clustering with UPARSE within USEARCH (Edgar, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2013\u003c/span\u003e) at 97% of similarity. The most abundant sequences were chosen as representatives for each Operational Taxonomic Unit (OTU), singletons were discarded, and OTU tables were rarefied to a common sampling depth of 1,583 OTUs/sample using the rarefy_even_depth() function in the \u003cem\u003ephyloseq\u003c/em\u003e R package (McMurdie and Holmes, \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). Species-level matches were determined by BLASTn against Silva v138.1 (Quast et al., \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). Hill numbers were used to calculate diversity indexes (observed richness\u0026thinsp;=\u0026thinsp;H0 and exponential of Shannon\u0026thinsp;=\u0026thinsp;eH\u0026rsquo;) using the DivPart() function in the \u003cem\u003eentropart\u003c/em\u003e R package (Marcon and Herault, \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). Finally, the 16S rRNA sequences of the MAGs were aligned and mapped to the 16S rRNA amplicon sequences using blastn (maximum e-value of 1 \u0026times; 10\u003csup\u003e\u0026minus;\u0026thinsp;5\u003c/sup\u003e) and bowtie2 v2.4.1 (Langmead and Salzberg, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2012\u003c/span\u003e), respectively. Alignment and mapping results were consistent for 45 out of the 48 MAGs possessing a 16S rRNA copy in their genome (\u003cb\u003eTable S8\u003c/b\u003e), and in agreement with the MAG GTDB taxonomy. In addition, the relative abundance of MAGs and their corresponding OTUs was correlated (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e2\u003c/span\u003eD).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eFunctional characterization of the deadwood microbial communities\u003c/h2\u003e \u003cp\u003eRelevant functions in wood decomposition were screened (Tl\u0026aacute;skal et al., \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2021a\u003c/span\u003e). All these analyses were performed on short reads, long reads and MAGs, and the results were compared. EukRep v0.6.7 (West et al., \u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) were used to predict contigs of eukaryotic origin. Metatranscriptomes were used to validate the expression of specific metabolisms.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003eCharacterization of CAZymes\u003c/h2\u003e \u003cp\u003eGenes involved in the decomposition of the easily degradable carbohydrates (arabinogalactanases, xylanases/xyloglucanases, mannanases, cellobiases and xylobiases), complex plant cell wall biopolymers (endoglucanase, cellobiohydrolase, exoglucanase), microbial biomass (peptidoglycansases, beta-glucanases, chitinases) and reserve compounds (alphaglucanases) were predicted using FragGeneScan v1.31 (Rho et al., \u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e2010\u003c/span\u003e), and annotated using run_dbcan.py (v2) program (Zhang et al., \u003cspan citationid=\"CR88\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Predicted prokaryotic and eukaryotic genes were compared to the dbCAN database V9 using HMMER 3.0 (Finn et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). See \u003cb\u003eTable S11\u003c/b\u003e for the details of the CAZymes screened for each activity. To validate Myxococcota cellulose decomposition activity, we mapped transcripts from sample 7 to Myxococcota MAGs (both Myxococcota genomes were assembled from this sample) using minimap2 v2.24 (Li, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) with the \u0026lsquo;-x sr\u0026rsquo; setting. Transcript counts were summed using dirseq v0.4.3 (Woodcroft et al., \u003cspan citationid=\"CR86\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) based on gene coordination from the gene prediction. The number of CAZyme for targeted activity were proportional between the short reads and the long reads (\u003cb\u003eTable S9\u003c/b\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003eInvestigating the genomic regions of nitrogen fixation\u003c/h2\u003e \u003cp\u003ePredicted genes from long reads and short reads co-assemblies were first screened with HMMER v3.3.2 against the KOfam database (accessed in June 2022) using the predefined adaptive thresholds (Aramaki et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). The number of metabolic pathways involved in the nitrogen cycle found in a single contig is reported in \u003cb\u003eTable S9\u003c/b\u003e. We further analyzed the contigs in which genomic region associated with N\u003csub\u003e2\u003c/sub\u003e fixation (i.e. nif genes\u0026thinsp;+\u0026thinsp;fix genes\u0026thinsp;+\u0026thinsp;Isc system genes or Suf system genes) were detected. We performed gene prediction using GeneMarkS v1.14 (Besemer, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2001\u003c/span\u003e) with default setting, and the annotation using eggnog-mapper v2.1.11 (Cantalapiedra et al., \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) and MMseqs after translating input CDS to proteins. Only contigs from PacBio HiFi co-assembly contained essential genes for nitrogen fixation, and five of them were considered as potentially functional (\u003cb\u003eTable S12\u003c/b\u003e). The phylogenetic relationships between the \u003cem\u003enifH\u003c/em\u003e sequences were analyzed using neighbor-joining trees generated using the maximum likelihood algorithm with 1,000 bootstrap iterations. First, multiple sequence alignment was generated using mafft v7.490 (Katoh et al., \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) with the option \u0026lsquo;--maxiterate 1000\u0026rsquo; and \u0026lsquo;--localpair\u0026rsquo;, then sequences were trimmed using BMGE v.2.0 (Criscuolo and Gribaldo, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2010\u003c/span\u003e) software and the BLOSUM30 matrix. Finally, the tree was computed using iqtree v2.2.0.3 (Nguyen et al., \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2015\u003c/span\u003e) and the LG\u0026thinsp;+\u0026thinsp;C20 model and visualized using FigTree v1.4.4 (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://tree.bio.ed.ac.uk/software/figtree/\u003c/span\u003e\u003cspan address=\"http://tree.bio.ed.ac.uk/software/figtree/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e).\u003c/span\u003e\u003c/p\u003e \u003cp\u003eOne of the nitrogen fixation contig (s509.ctg000513l) was also found in the 36_Steroidobacteraceae MAG, a gammaproteobacterial genome assembled in this study. To investigate nitrogen fixation potential in the Steroidobacteraceae family, we investigated 71 Steroidobacteraceae genomes from the GEMs catalog (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://portal.nersc.gov/GEM/\u003c/span\u003e\u003cspan address=\"https://portal.nersc.gov/GEM/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) together with 110 isolated strain genomes of Steroidobacteraceae from NCBI. We annotated these genomes using eggnog-mapper v2.1.11, with the option \u0026lsquo;--itype metagenome\u0026rsquo; for gene prediction from Diamond/MMseqs2 blastx hits, and no nitrogenase genes were identified (\u003cb\u003eTable S12\u003c/b\u003e). Given the transferability of the essential genes for nitrogen fixation (Bolhuis et al., \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2010\u003c/span\u003e), we studied genomic islands of 36_Steroidobacteraceae with Islandviewer4 (Bertelli et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) after having identified the start position of the genome with circlator v1.5.5 (Hunt et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2015\u003c/span\u003e), and plotted the results using Proksee (Grant et al., \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). We found no evidence of nitrogen fixation capability acquired from horizontal gene transfer (HGT) in 36_Steroidobacteraceae. The genomic islands corresponded well to variation of GC content (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e4\u003c/span\u003eC), but occurred neither inside nor around the s509.ctg000513l contig (located between 5118742 and 5335241 bp). We further studied phylogenetic similarities of 36_Steroidobacteraceae \u003cem\u003enifH\u003c/em\u003e sequences in the Interpro database (accessed on August 1st, 2023). First, the \u003cem\u003enifH\u003c/em\u003e sequences were clustered at 97% of similarity using USEARCH v11.0.667. Then, multiple sequence alignment was generated using mafft v7.490 with the option \u0026lsquo;--maxiterate 1000\u0026rsquo; and \u0026lsquo;--localpair\u0026rsquo;. We trimmed the alignment using BMGE v.2.0 software and the BLOSUM30 matrix. We computed the tree using iqtree v2.2.0.3 and the LG\u0026thinsp;+\u0026thinsp;C20 model and rooted the tree using minimal ancestor deviation (Tria et al., \u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). We next sub-selected the closest related taxa and used FigTree v1.4.4 for visualization. Finally, we confirmed the transcription of the nitrogen fixation genes in 36_Steroidobacteraceae by mapping the sample 7 transcripts (36_Steroidobacteraceae genomes were assembled from this sample) to s509.ctg000513l contig using minimap2 v2.24 with \u0026lsquo;-x sr\u0026rsquo; setting. Transcript counts were summed using dirseq v0.4.3 (Woodcroft et al., \u003cspan citationid=\"CR86\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) based on gene coordination from the gene prediction. We also confirmed the nitrogen fixation gene expression in the five PacBio HiFi contigs by mapping the sample 7 transcript to nitrogenase genes (\u003cem\u003enifH\u003c/em\u003e, \u003cem\u003enifD\u003c/em\u003e, \u003cem\u003enifK\u003c/em\u003e) to make the results comparable (\u003cb\u003eTable S12\u003c/b\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eBiosynthetic Gene Clusters\u003c/h2\u003e \u003cp\u003eAntiSMASH v6.1.1 (Blin et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) was used to predict secondary metabolite clusters. BGCs were predicted with \u0026lsquo;--hmmdetection-strictness strict\u0026rsquo; and \u0026lsquo;--asf\u0026rsquo; settings. We also added the options \u0026lsquo;--cb-general\u0026lsquo;, \u0026lsquo;--cb-subclusters\u0026lsquo; and \u0026lsquo;--cb-knownclusters\u0026lsquo; in order to compare identified clusters against antiSMASH-predicted clusters database, known subclusters responsible for synthesizing precursors and known gene clusters from the MIBiG database (Terlouw et al., \u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) respectively. To cluster the BGCs by similarity, we used BiG-SCAPE v1.1.5 (Navarro-Mu\u0026ntilde;oz et al., \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) with the \u0026lsquo;--mibig\u0026lsquo; and \u0026lsquo;--cutoffs 0.6\u0026lsquo; flags to cluster the BGC detected and the MIBiG database v3.1 entries at 60% percent of similarity.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003ePacBio HiFi and Illumina sequence assembly and binning\u003c/h2\u003e \u003cp\u003ePacBio HiFi and Illumina HiSeq sequencing platforms generated respectively 16.1 Gb and 19.8 Gb for the four deadwood samples (\u003cb\u003eTable \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e\u003c/b\u003e). For the samples 6, 7, and 57, PacBio HiFi sequencing generated on average 282,091 sequences per sample, with an average sum length of 5.2 Gb. For the sample 84, PacBio HiFi sequencing produced 19,734 sequences with an average sum length of 0.3 Gb. Illumina sequencing yielded on average 21,168,790 sequences per sample, with an average sum length of 4.9 Gb (\u003cb\u003eTable \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e\u003c/b\u003e). The average sequence length from PacBio HiFi was approximately 18,300 bp and 234 bp from Illumina. Sequences affiliated to Prokaryotes (almost exclusively Bacteria) dominated in both PacBio HiFi raw reads (85%) and Illumina reads (51%), while the share of sequences affiliated to Eukaryotes was 13% for PacBio HiFi raw reads and 19% for Illumina raw reads (\u003cb\u003eTable \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e\u003c/b\u003e). About 30% Illumina raw reads remained unclassified.\u003c/p\u003e \u003cp\u003eThe PacBio HiFi sequences were co-assembled into 16,446 contigs with an N50 of 66,188 bp, with 16,439 contigs\u0026thinsp;\u0026gt;\u0026thinsp;10 kb and 9,696 contigs\u0026thinsp;\u0026gt;\u0026thinsp;50 kb (\u003cb\u003eTable S3\u003c/b\u003e). The largest contig reached a size of 5.8 Mb. The Illumina short reads co-assembly generated 3,734,801 contigs of which 102,279 contigs were \u0026gt;\u0026thinsp;1 kb, and 109 were \u0026gt;\u0026thinsp;10 kb, with an N50 of 685 bp (\u003cb\u003eTable S3\u003c/b\u003e). The largest contig had a size of 0.1 Mb. The total length of assembled long-read contigs was 1.2 Gb, while for short-read contigs was around half that (i.e. 0.6 Gb). The average proportion of long and short read mapping to contigs was 41% and 39%, respectively (\u003cb\u003eTable S4\u003c/b\u003e). See \u003cb\u003eTable S5\u003c/b\u003e for details on sample-by-sample assembly for PacBio HiFi and Illumina HiSeq.\u0026nbsp;Contigs affiliated to Prokaryotes dominated sample-by-sample assemblies and co-assemblies from both PacBio HiFi and Illumina data. Proportions of contigs affiliated to eukaryota were consistent between sample-by-sample assembly and co-assembly, accounting respectively for 5.5% and 5.1% of short reads and 8.5% and 10.5% of long reads (\u003cb\u003eTable \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e\u003c/b\u003e). K-mer-based analysis supported these results, identifying 3,168 eukaryotic contigs (270 kb) in the PacBio HiFi co-assembly and 454 eukaryotic contigs (2 kb) in the PacBio HiFi co-assembly.\u003c/p\u003e \u003cp\u003eTwenty one percent of the raw PacBio HiFi reads were successfully binned (\u003cb\u003eTable S6\u003c/b\u003e), constituting a total of 69 unique bacterial metagenome-assembled genomes (MAGs). Among these, 14 were High-quality draft MAGs (i.e. completeness\u0026thinsp;\u0026gt;\u0026thinsp;90% and contamination\u0026thinsp;\u0026lt;\u0026thinsp;5%, presence of the 23S, 16S, and 5S rRNA genes and at least 18 tRNAs, \u003cb\u003eTable S6\u003c/b\u003e) and 55 were Medium quality draft MAGs (completeness\u0026thinsp;\u0026gt;\u0026thinsp;50%, contamination\u0026thinsp;\u0026lt;\u0026thinsp;10%, \u003cb\u003eFigure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e\u003c/b\u003e). Twenty-two MAGs were composed of less than 10 contigs, among which 7 were composed of a single contig and considered nearly finished MAGs (\u003cb\u003eTable S6\u003c/b\u003e). Ribosomal RNA (i.e., 16S, 23S and/or 5S rRNA) genes were found in 53 of the 69 MAGs (\u003cb\u003eTable S6\u003c/b\u003e). Eleven unique MAGs were recovered from the Illumina HiSeq co-assembly and sample-by-sample assembly, all of them Medium-quality draft MAGs (\u003cb\u003eTable S7, Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e\u003c/b\u003e). These MAGs consisted of 772\u0026thinsp;\u0026plusmn;\u0026thinsp;603 contigs, with an average length of 3,480\u0026thinsp;\u0026plusmn;\u0026thinsp;4,254 bp, while the MAGs generated from PacBio HiFi assemblies consisted of 26\u0026thinsp;\u0026plusmn;\u0026thinsp;21 contigs with an average length of 148,129\u0026thinsp;\u0026plusmn;\u0026thinsp;386,395 bp. Six of the 11 MAGs obtained from the short-read assemblies were also recovered from long-read data.\u003c/p\u003e \u003cp\u003e \u003cb\u003eIntegrating\u003c/b\u003e \u003cb\u003ede novo\u003c/b\u003e \u003cb\u003egenome assembly and metabarcoding\u003c/b\u003e\u003c/p\u003e \u003cp\u003eBased on metabarcoding, prokaryotic 16S sequences in the deadwood were most frequently assigned to Proteobacteria (43%), and to a lesser extent to Acidobacteriota (14%), Actinobacteriota (10%), Bacteroidota (9%), Verrucomicrobiota (9%), Planctomycetota (8%), Myxococcota (3%) and Patescibacteria (1%) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e2\u003c/span\u003eA). MAGs from seven of these eight bacterial phyla with a relative abundance\u0026thinsp;\u0026gt;\u0026thinsp;1% identified by 16S metabarcoding were successfully binned from PacBio HiFi assemblies. This included 38 genomes belonging to Proteobacteria (31 to Alphaproteobacteria and 7 to Gammaproteobacteria), 10 to Bacteroidota, 6 to Verrucomicrobiota, 6 to Patescibacteria, 4 to Acidobacteriota, 3 to Actinobacteriota, and 2 to Myxococcota (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e2\u003c/span\u003eB). The number of MAGs recovered per phylum showed positive correlation with the OTU richness of this phylum (Pearson correlation coefficient, cor\u0026thinsp;=\u0026thinsp;0.84, p-value\u0026thinsp;=\u0026thinsp;0.008). However, despite Planctomycetota contigs were recovered (\u003cb\u003eFigure \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e\u003c/b\u003e), no Planctomycetota genomes were assembled, probably due to the specific community structure of this clade. Planctomycetota exhibited a relatively low 16S rRNA abundance compared to the seven other most abundant phyla (ranking as the sixth most abundant phylum) but were the second prokaryotic phylum after Proteobacteria in terms of OTU richness and diversity (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e2\u003c/span\u003eA).\u003c/p\u003e \u003cp\u003eAmong the 48 MAGs containing a copy of the 16S rRNA gene, 44 MAGs were successfully linked to a unique OTU recovered using 16S metabarcoding. In the remaining four cases, two different MAGs aligned with a single OTU (\u003cb\u003eTable S8\u003c/b\u003e), likely due to a low resolution (e.g. size) of the 16S rRNA gene fragment. Within these 48 MAGs, 24 were associated with OTUs exhibiting a relative abundance of 16S greater than 0.1% (i.e., among the 172 OTUs, Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e2\u003c/span\u003eC), including 5 MAGs originating from the 12 most abundant OTUs. Of these, 3 MAGs corresponded to Alphaproteobacteria and 2 corresponded to Myxoccocota. Additionally, 20 MAGs corresponded to OTUs with a relative abundance, based on metabarcoding, below 0.1%. This included two assigned to Patescibacteria and two assigned to Alphaproteobacteria. While the relative abundance of the MAGs where generally proportional to the relative abundance of the respective OTU (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e2\u003c/span\u003eD), one abundant Patescibacteria MAG (5_UBA9983_A) had a low 16S relative abundance (\u003cb\u003eTable S6\u003c/b\u003e, \u003cb\u003eTable S8\u003c/b\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eCAZymes encoded by the deadwood microorganisms\u003c/h2\u003e \u003cp\u003eThe abundance of CAZymes for each activity was comparable across short reads, long reads, and MAGs (\u003cb\u003eFigure S3\u003c/b\u003e). In the short-read co-assembly, we identified a total of 7,617 CAZymes, while 5,161 CAZymes were found in the long-read co-assembly. Among these, alphaglucanases, peptidoglycanases, cellobiases, and xylobiases were the most abundant enzymes, whereas exoglucanases and cellobiohydrolases were relatively rare. Notably, the majority of identified CAZymes were bacterial in origin. Short-read data did not yield any eukaryotic CAZymes, whereas 197 eukaryotic CAZymes were identified in the PacBio HiFi long-read data. When considering the total gene count in the long-read co-assembly, the proportion of CAZymes represented 0.37% for bacteria and 0.04% for eukaryotes (\u003cb\u003eTable S9\u003c/b\u003e).\u003c/p\u003e \u003cp\u003eThe decomposition of easily decomposable OM mediated by cellobiases and xylobiases was frequently observed in the recovered genomes (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e3\u003c/span\u003e), with only 3 MAGs lacking this capability. Additionally, CAZymes targeting reserve compounds and microbial biomass were abundant across most MAGs, although relatively rare in Patescibacteria and a few other MAGs. Notably, cellulose and chitin decomposition capabilities were less common, present in 52% and 51% of the genomes, respectively. MAGs affiliated with Myxococcota (Polyangiaceae family) encoded the highest number of CAZymes involved in the degradation of both readily decomposable and recalcitrant biopolymers, including those targeting reserve compounds. However, they lacked enzymes targeting fungal biomass decomposition (chitinases). Importantly, Myxococcota MAGs exhibited 14 times more cellulose-targeting enzymes than other MAGs in this study and possessed the set of all necessary enzymes for cellulose decomposition (endoglucanases, exoglucanases, and cellobiohydrolases). Transcript mapping revealed that endoglucanases were commonly transcribed by all prokaryotic MAGs where they were present (\u003cb\u003eTable S10\u003c/b\u003e). However, exoglucanases were predominantly transcribed by Polyangiaceae (Myxococcota) MAGs, and cellobiohydrolases were exclusively transcribed by these MAGs. Bacteroidetes exhibited a high proportion of CAZymes targeting microbial biomass, along with numerous enzymes targeting easily degradable carbohydrates and a few targeting reserve compounds. Conversely, Acidobacteriota and Verrucomicrobiota showed proportionally fewer enzymes targeting microbial biomass but many involved in the decomposition of easily degradable carbohydrates and reserve compounds. Proteobacteria encoded numerous CAZymes targeting microbial biomass, with an uneven distribution of enzymes involved in easily degradable carbohydrates and reserve compound decomposition, with some encoding few and others encoding many (\u003cb\u003eTable S10\u003c/b\u003e).\u003c/p\u003e \u003cp\u003eThe number of CAZymes found in MAGs was proportional to their genome size (cor\u0026thinsp;=\u0026thinsp;0.76, p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Consequently, the phyla encoding the smallest number of CAZymes were Actinobacteriota and Patescibacteria. Actinobacteriota assembled in deadwood (Nanopelagicaceae and Microbacteriaceae families), encoded mainly CAZymes targeting reserve compounds, had limited number of CAZymes targeting easily degradable carbohydrates, and lacked the metabolic capacity to degrade complex biopolymers of plant cell wall and microbial biomass (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Patescibacteria exhibited only a few enzymes involved in decomposition of easily degradable carbohydrates, targeting microbial biomass and reserve compounds. However, transcript mapping confirmed expression of CAZymes by Patescibacteria, exceeding the transcription of Acidobacteriota and Proteobacteria for mannanase and betaglucanases activity (\u003cb\u003eTable S10\u003c/b\u003e). Importantly, while the level of chitinase expression was generally low among recovered MAGs (mean coverage\u0026thinsp;=\u0026thinsp;3.0), Patescibacteria were the second most active phylum in terms of chitinase expression after Bacteroidota (6.6 and 7.3 respectively).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eNitrogen cycling metabolism\u003c/h2\u003e \u003cp\u003eThe number of metabolic pathways involved in the nitrogen cycle was low. Of those identified, dissimilatory and assimilatory nitrate reduction were the most frequent in Pacbio HiFi contigs, followed by nitrogen fixation \u003cb\u003e(Table S9)\u003c/b\u003e. In addition, a contig with the ammonia methane/monooxygenase (\u003cem\u003epmo\u003c/em\u003e-\u003cem\u003eamo\u003c/em\u003e) ABC subunit was identified, suggesting a potential for nitrification. Consistently, genes coding for N\u003csub\u003e2\u003c/sub\u003e fixation and dissimilatory or assimilatory nitrate reduction were present in 14 MAGs (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e3\u003c/span\u003e). The genes \u003cem\u003enirB\u003c/em\u003e and \u003cem\u003enirD\u003c/em\u003e, converting nitrite to ammonia in the second step of dissimilatory nitrate reduction pathway, were found in Proteobacteria (3 Alphaproteobacteria MAGs), Myxococcota (2 Polyangiales MAGs) and Acidobacteriota (Bryobacterales). The gene \u003cem\u003enirA\u003c/em\u003e, catalyzing the conversion of nitrite to ammonia in the second step of assimilatory nitrate reduction pathway, was found in Verrucomicrobiota (Methylacidiphilales) and Alphaproteobacteria (3 Rhizobiales MAGs). The genes \u003cem\u003enifH\u003c/em\u003e, \u003cem\u003enifD\u003c/em\u003e, and \u003cem\u003enifK\u003c/em\u003e (i.e. nitrogenase genes), responsible for the biological reduction of dinitrogen to ammonia, were only found in one Gammaproteobacteria MAG belonging to the Steroidobacterales.\u003c/p\u003e \u003cp\u003eTo gain deeper insights into nitrogen fixation during wood decomposition processes, we examined the genomic regions associated with N\u003csub\u003e2\u003c/sub\u003e fixation in both short-read and long-read contigs. No contigs generated by the short-read co-assembly contained essential genes for nitrogen fixation, whereas the PacBio HiFi co-assembly revealed five contigs with genomic regions associated to N\u003csub\u003e2\u003c/sub\u003e fixation (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e4\u003c/span\u003eA). All nitrogenase genes identified in these contigs were transcribed (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e4\u003c/span\u003eB), suggesting that they are functional during \u003cem\u003ein situ\u003c/em\u003e nitrogen fixation. The minimum set of genes encoding structural and biosynthetic components \u0026mdash; specifically NifHDK and NifENB \u0026mdash; were present. The \u003cem\u003efixABCRUX\u003c/em\u003e genes were also present, preferred to the \u003cem\u003eRnf\u003c/em\u003e complex to catalyze the production of reduced ferredoxin/flavodoxin. Four of these contigs were affiliated to Alphaproteobacteria and one to Gammaproteobacteria (\u003cb\u003eTable S12\u003c/b\u003e). The Gammaproteobacteria contig lacked \u003cem\u003enifS\u003c/em\u003e but encoded the complete \u003cem\u003eIsc\u003c/em\u003e system (\u003cem\u003eiscR\u003c/em\u003e, \u003cem\u003eiscS\u003c/em\u003e, \u003cem\u003eiscU\u003c/em\u003e, \u003cem\u003eiscA\u003c/em\u003e, \u003cem\u003ehscB\u003c/em\u003e, \u003cem\u003ehscA\u003c/em\u003e, \u003cem\u003efdx\u003c/em\u003e, and \u003cem\u003eiscX\u003c/em\u003e), required for maturation of [Fe-S] proteins. This system was absent in the Alphaproteobacteria contigs, where it was substituted by the \u003cem\u003eSuf\u003c/em\u003e system (\u003cem\u003esufB\u003c/em\u003e, \u003cem\u003esufC\u003c/em\u003e, \u003cem\u003esufD, sufS\u003c/em\u003e, and \u003cem\u003esufE\u003c/em\u003e).\u003c/p\u003e \u003cp\u003eThe gammaproteobacterial nitrogen fixation contig (s509.ctg000513l) was binned into the 36_Steroidobacteraceae MAG. Transcript mapping against this MAG validated the expression of essential genes for nitrogen fixation (\u003cb\u003eTable S12\u003c/b\u003e). The identification of nitrogen fixation genes within the Steroidobacteraceae family is noteworthy. Despite annotating 181 Steroidobacteraceae genomes from the Earth's Microbiomes catalog and the NCBI database (71 and 110 respectively, \u003cb\u003eTable S13\u003c/b\u003e), revealing individual genes potentially implicated in nitrogen fixation (e.g., \u003cem\u003efixB\u003c/em\u003e, \u003cem\u003eiscASX\u003c/em\u003e, \u003cem\u003enifZ\u003c/em\u003e, \u003cem\u003esufBCES\u003c/em\u003e, \u003cem\u003efdx\u003c/em\u003e), none of the annotated genomes encompassed the complete set of essential nitrogen fixation genes (\u003cb\u003eTable S12\u003c/b\u003e). Moreover, extensive investigation of the NCBI nr and Interpro databases (\u003cb\u003eTable S12\u003c/b\u003e) failed to identify \u003cem\u003enifH\u003c/em\u003e in the Steroidobacteraceae family. Phylogenetic analyses of 36_Steroidobacteraceae \u003cem\u003enifH\u003c/em\u003e and the entire s509.ctg000513l contig indicated similarities with the Methyloccocales order (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e4\u003c/span\u003eD, \u003cb\u003eTable S12\u003c/b\u003e), yet we found no evidence supporting recent HGT of this region into the Steroidobacteriaceae MAG (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e4\u003c/span\u003eC, see \u003cb\u003eMethods\u003c/b\u003e for details). Our analyses also found no support for mis-binning, affirming the accurate assembly of 36_Steroidobacteraceae (\u003cb\u003eTable S6\u003c/b\u003e). Collectively, these data underscore a functional potential for nitrogen fixation within this deadwood-associated member of the Steroidobacteraceae family.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eBiosynthetic Gene Clusters in deadwood microorganisms\u003c/h2\u003e \u003cp\u003eA total of 1,089 putative Biosynthetic Gene Clusters (BGCs) were identified in the PacBio HiFi long-read co-assembly, presenting a stark contrast to the three BGCs found in the short-read co-assembly (\u003cb\u003eFigure S4\u003c/b\u003e). The most frequent BGC types were terpenes (291) and ribosomally synthesized and post-translationally modified peptides (RiPPs, 235). While the majority of BGCs were located on bacterial contigs, we also uncovered eukaryotic BGCs, including terpenes, non-ribosomally synthesized peptides (NRPS), polyketides, and siderophores (\u003cb\u003eFigure S4\u003c/b\u003e). A comparison with the MIBig database highlighted the novelty of the identified BGCs, with a staggering 98% showing less than 50% shared genes with MIBig entries.\u003c/p\u003e \u003cp\u003eThe primary BGC types identified in the long-read assembly were also found in MAGs (\u003cb\u003eFigure S4\u003c/b\u003e), enabling the association of biosynthetic potential with taxonomic groups. Proteobacteria, particularly MAGs from Rhizomicrobium, Rhizobiales (Alphaproteobacteria), and Steroidibacteriaceae (Gammaproteobacteria), carried the majority of BGCs, exceeding 10 in some instances (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e5\u003c/span\u003eA). Proteobacterial BGCs presented a wide diversity, encompassing various types of compounds such as thioamitides, lassopeptides, ranthipeptides, azol(in)e-containing linear peptides, linaridines, or ladderanes, each holding bioprospecting potential. The two Myxococcota MAGs displayed the highest numbers of BGCs (14 and 15), encoding a wide array of compounds, including thioamitides and ranthipeptides. Verrucomicrobiota, Acidobacteriota, Bacteroidota, and Actinobacteriota MAGs harbored 0\u0026ndash;8 BGCs, with terpene, RiPP, polyketide, NRPS, and arylpolyene BGCs being the most prevalent. Group-specific BGCs, such as flexirubins in Chitinophagaceae (Bacteroidota) MAGs, were also identified. Patescibacteria MAGs did not contain any BGCs. BiG-SCAPE analysis revealed that the 271 BGCs identified in MAGs were categorized into 254 unique gene cluster families (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e5\u003c/span\u003eB). The abundance of singletons underscored the vast diversity of BGCs in deadwood MAGs. For instance, seven BGCs encoding thioamitides each belonged to a distinct gene cluster family (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e5\u003c/span\u003eC). BGCs belonging to one cluster family always belonged to taxonomically related MAGs, indicating that BGCs in deadwood bacteria are phylogenetically conserved.\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this study, we compared the performance of PacBio HiFi and Illumina sequencing platforms and tested different methods of assembly for metagenomes of deadwood samples. The two platforms generated roughly similar amounts of data, but the PacBio HiFi sequencing produced much longer reads than Illumina. Hifiasm-meta produced the best assemblies for PacBio HiFi data, yielding twice the amount of data and longer contigs than Illumina megahit assemblies. Hybrid assembly was more computationally demanding, and did not improve the number and size of contigs. This is probably because hybrid assembly was originally developed to improve short-read assembly (Wick et al., \u003cspan citationid=\"CR84\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), as the sequencing error generated by long-read sequencing was high. However, the repetitive library which calls for consensus reads developed by the PacBio HiFi sequencing approach, has significantly improved nucleotide accuracy (Marx, \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). As a result, short reads are no longer needed to improve the assembly of long reads if sequenced by PacBio HiFi.\u003c/p\u003e \u003cp\u003ePacBio HiFi metagenomic sequencing has demonstrated outstanding efficiency for reconstructing the genomes from deadwood microbial communities, including bacteria that could not be previously cultivated from these samples (Tl\u0026aacute;skal et al, 2021) or elsewhere. A total of 69 bacterial genomes were generated from 16 Gb of PacBio HiFi reads from 4 deadwood samples, including 67 newly reconstructed MAGs, 14 high-quality and 7 composed of a single contig. It outperformed Illumina assemblies by 6.3 times (11 MAGs), producing less fragmented, less contaminated and more complete genomes. In comparison to related studies, our findings overtake those of (Tl\u0026aacute;skal et al., \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2021b\u003c/span\u003e), who assembled 58 MAGs from 25 short-read metagenomes of deadwood from the same forest. Our investigation extended to the eight most abundant deadwood bacterial phyla, revealing successful genome assembly for all phyla except Planctomycetota. Our results suggest that while genome size and relative abundance may not be the ultimate limiting factors for successful binning, the inherent complexity of Planctomycetota, characterized by high species richness and phylogenetic diversity (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e2\u003c/span\u003eA), challenges the recovery of their genomes.\u003c/p\u003e \u003cp\u003eIllumina and PacBio HiFi sequencing generated a proportional number of eukaryotic sequences, but PacBio HiFi assemblies yielded more eukaryotic contigs than Illumina assemblies. This discrepancy is likely attributed to the longer sequencing lengths and high accuracy of PacBio HiFi sequencing (Uliano-Silva et al., \u003cspan citationid=\"CR78\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Eukaryotic genomes, characterized by intricate features like repetitive regions, introns, and exons (Galagan et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2005\u003c/span\u003e), require greater metagenomic sequencing depth and sample size to be successfully assembled compared to prokaryotic genomes. For instance, (Saraiva et al., \u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) needed 6,000 terrestrial metagenomes to assemble 197 eukaryotic bins, whereas (Ma et al., \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) achieved the assembly of 40,039 prokaryotic MAGs from 2,941 soil metagenomes.\u003c/p\u003e \u003cp\u003eEmploying PacBio HiFi sequencing, we unveiled the important role of Myxococcota in deadwood decomposition. While Myxococcota have been previously assembled from various ecosystems, including aquatic, terrestrial, host-associated, and built environments (Nayfach et al., 2021), our study marks the first assembly of Myxococcota genomes within the deadwood ecosystem. Characterized by relatively large genomes (8 Mbp according to GTDB database r202), Myxococcota exhibit unique traits, encompassing motility (Nan et al., \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2013\u003c/span\u003e), predation (Thiery and Kaimer, \u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), fruiting bodies formation (Mu\u0026ntilde;oz-Dorado et al., \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2016\u003c/span\u003e), and the ability to decompose cellulose (L\u0026oacute;pez-Mond\u0026eacute;jar et al., \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Here, we successfully assembled two abundant Myxococcota (Polyangiaceae) genomes expressing high cellulosic activity, including endoglucanase, exoglucanase, and cellobiohydrolase \u0026ndash; enzymes exclusively found in Myxococcota MAGs (\u003cb\u003eTable S10\u003c/b\u003e). While Polyangiaceae also expressed CAZymes targeting bacterial biomass, the lack of chitinase prevented them from recycling fungal biomass. Conversely, chitinases were consistently present in Bacteroidota and detected in Verrucomicrobiota, Proteobacteria and Patescibacteria. Remarkably, the sole chitinase found in Patescibacteria 5_UBA9983_A demonstrated the second-highest level of expression (\u003cb\u003eTable S10\u003c/b\u003e).\u003c/p\u003e \u003cp\u003ePatescibacteria had been identified in a wide range of ecosystems (Nayfach et al., 2021) including deadwood habitats (Choi et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Tl\u0026aacute;skal et al., \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2021b\u003c/span\u003e), yet their ecological role remains enigmatic. While their limited genome size, lacking crucial metabolic genes, suggests an obligate epibiotic lifestyle (Kuroda et al., \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), their parasitic status remains subject to debate (Wang et al., \u003cspan citationid=\"CR81\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). In our study, UBA9983_A demonstrated a specialized opportunistic lifestyle, exclusively recycling fungal cell wall components (\u003cb\u003eTable S10\u003c/b\u003e). Additionally, Patescibacteria of the order Sacharrimonadales showcased elevated mannase expression, an enzyme involved in hemicellulose degradation, highlighting their active contribution in wood decomposition processes. These findings collectively imply that Patescibacteria may not solely depend on host resources. Further analyses are needed to elucidate whether hemicellulose- and fungal-derived residues serve as energy sources for ATP generation or benefits to the host (e.g., commensalism or mutualism). Nevertheless, our results suggest a broader ecological relationship for Patescibacteria beyond parasitism.\u003c/p\u003e \u003cp\u003eNitrogen concentration in deadwood is low (Tl\u0026aacute;skal et al., \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2021a\u003c/span\u003e) and probably represents a limiting factor for efficient decomposition. As metabolic pathways involved in nitrogen assimilation were scarce in the studied samples (\u003cb\u003esee Results\u003c/b\u003e), microbial biomass could represent an alternative source of nitrogen (L\u0026oacute;pez-Mond\u0026eacute;jar et al., \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). We found that CAZymes targeting bacterial biomass were more frequent than CAZymes targeting fungal biomass, but that peptidoglycanases were 2 times less expressed than chitinases (\u003cb\u003eTable S11\u003c/b\u003e, \u003cb\u003eTable S10\u003c/b\u003e). This result is surprising since fungal biomass contains less nitrogen than bacterial biomass (Paul and Frey, \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), but these processes might simply be regulated by the availability of bacterial and eukaryotic biomass. While microbial biomass decomposition represents an interesting strategy for recycling nitrogen during decay, an external source of nitrogen might be required to initiate decomposition of fresh dead wood. The initial nitrogen input is likely provided by bacterial nitrogen fixation, as illustrated by nitrogen fixation rates being eight times higher in young deadwood compared to old deadwood (Tl\u0026aacute;skal et al., \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2021a\u003c/span\u003e). The conversion of atmospheric N\u003csub\u003e2\u003c/sub\u003e to NH\u003csub\u003e3\u003c/sub\u003e, being a highly energy-consuming process (Cherkasov et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2015\u003c/span\u003e), is only realized if no better suitable nitrogen sources are available (Burris and Roberts, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e1993\u003c/span\u003e). The continuous expression of nitrogenase genes in later stages of decomposition (\u0026gt;\u0026thinsp;4 years) thus indicates that the recycling of microbial biomass does not completely meet the microbial nitrogen demand throughout the decomposition process.\u003c/p\u003e \u003cp\u003eDespite biological nitrogen fixation has been extensively investigated (e.g. Davies-Barnard and Friedlingstein, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Dos Santos et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Zehr and Capone, \u003cspan citationid=\"CR87\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), our study marks the first recovery of this function within the Steroidobacteraceae family. Beyond reporting a novel nitrogen-supplying bacteria in a nitrogen-limited environment, PacBio HiFi sequencing facilitated the investigation of genes involved in the conversion of N\u003csub\u003e2\u003c/sub\u003e into ammonium. In the deadwood ecosystem, Proteobacteria catalyze the production of reduced ferredoxin/flavodoxin using the fix operon, particularly advantageous under oxygen-limited conditions (Alleman et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). In addition, Steroidobacteraceae employ the \u003cem\u003eisc\u003c/em\u003e system for biosynthesizing [Fe-S] proteins, crucial under elevated oxygen conditions (Johnson et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2006\u003c/span\u003e). The coexistence of the \u003cem\u003efix\u003c/em\u003e operon and the \u003cem\u003eisc\u003c/em\u003e system likely allows Steroidobacteraceae to maintain nitrogen fixation activity under oxygen concentration fluctuations. Alphaproteobacteria appear less sensitive to elevated oxygen, favoring the \u003cem\u003esuf\u003c/em\u003e system over the \u003cem\u003eisc\u003c/em\u003e system, which is known to be more beneficial for bacterial growth in the presence of hydrogen peroxide (Tokumoto, \u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e2004\u003c/span\u003e). Although the proportional contribution of Steroidobacteraceae (Gammaproteobacteria) and Alphaproteobacteria to global deadwood nitrogen fixation requires further exploration, our study provides empirical evidence of their \u003cem\u003ein situ\u003c/em\u003e activity.\u003c/p\u003e \u003cp\u003ePacBio HiFi sequencing unveiled the remarkable potential of deadwood microorganisms for the production of diverse secondary metabolites, an important trait that would be totally neglected if one would just rely on short read-based approaches. Our study identified over a thousand mostly novel BGCs, showcasing the extensive diversity of these BGCs in MAGs. This diversity suggests that deadwood bacteria display multiple interactions with other microorganisms in deadwood. Furthermore, our investigation highlighted bacterial groups in deadwood that hold promise for the production of novel bioactive compounds. Similar to observations in soil (Sharrar et al., \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), the abundance of BGC types varied by taxonomy. In deadwood, we observed high numbers of BGCs in Myxococcota and Proteobacteria, as well as in other groups such as Verrucomicrobiota and Acidobacteriota, whose biosynthetic potential has only recently been reported (Crits-Christoph et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Waschulin et al., \u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Notably, we did not identify a shared BGC family across different bacterial taxa, indicative of a compound with broader importance in deadwood. In contrast, taxonomic conservation of BGC families in deadwood MAGs was observed. The presence of BGCs for flexirubins, pigments typical of Bacteroidota (Brinkmann et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), in several Chitinophagaceae MAGs constitutes convincing proof of validity of our metagenome analysis.\u003c/p\u003e \u003cp\u003eLong-read sequencing-based analysis of metagenomes, while powerful, still has limitations. More effort would be needed to appropriately describe the fungal component of the deadwood microbiome and this is partly true also for the Planctomycetes: although their sequences have been obtained by PacBio HiFi sequencing, the high species richness and phylogenetic diversity of this phylum challenged their genome recovery. Still, long-read sequencing-based metagenome analysis stands unparalleled indicating microbiome functions not recoverable using the conventional approaches \u0026ndash; short-read metagenomics and culturing. It enabled the assembly of novel genomes of bacteria with key roles in deadwood decomposition: cellulose decomposition in Polyangiaceae, nitrogen fixation in Steroidobacteraceae. It also revealed significant contribution of Patescibacteria to wood decomposition processes, and identified the wealth of new BGCs with potential ecological and/or biotechnological significance.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003eAvailability of data and material\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDescriptions of the deadwood samples 6, 7, 57 and 84 are available at the NCBI BioSample repository (https://www.ncbi.nlm.nih.gov/biosample/), under accession numbers SAMN13925154, SAMN13925155, SAMN13925167, and SAMN13925168, respectively. The raw PacBio HiFi sequences from the four deadwood samples are available at the NCBI Sequence Read Archive repository (https://www.ncbi.nlm.nih.gov/sra/), under accession numbers SRR28211698 ̶ SRR28211701. The co-assembly of PacBio HiFi sequences is available at NCBI GenBank (https://www.ncbi.nlm.nih.gov/nuccore/) under accession number JBBCBH000000000. The 69 metagenome-assembled genomes from PacBio HiFi sequences are available at the NCBI GenBank repository, under accession numbers JBBCFX000000000 ̶ JBBCIN000000000. The raw Illumina HiSeq metagenome sequences corresponding to samples 6, 7, 57 and 84 are available at the NCBI Sequence Read Archive repository, under accession numbers SRR10968229, SRR10968228, SRR10968259, and SRR10968258, respectively. The Illumina HiSeq metatranscriptome sequences of samples 6 and 7 are available at the NCBI Sequence Read Archive repository, under accession numbers SRR10968251 and SRR10968250. The Illumina MiSeq 16S rRNA amplicon sequencing data for samples 6, 7, 57 and 84 are available at the NCBI Sequence Read Archive repository, under accession numbers SRR12914735, SRR12914734, SRR12914771, and SRR12914770. The biosynthetic gene clusters recovered from Illumina short read metagenome, PacBio HiFi long read metagenome and from PacBio HiFi MAGs, are available at the Zenodo repository (https://zenodo.org), under record numbers 10550953, 10529179, and 10529038, respectively.\u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by the Ministry of Education, Youth and Sports of the Czech Republic (CZ.02.01.01/00/22_008/0004635 - AdAgriF - Advanced methods of greenhouse gases emission reduction and sequestration in agriculture and forest landscape for climate change mitigation).\u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026apos; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eM.K. and P.B. designed the study. E.R., M.K. and P.B. directed the analyses. M.K. and V.T. performed HMW DNA isolation. P.T.D. and E.R. carried out the computational analyses and visual representations. E.R and M.K. analyzed the data with the help of P.T.D., V.T., R.L.M. and P.B. E.R and M.K. wrote the original manuscript, with the input of P.T.D., V.T., R.L.M. and P.B. All authors have read and approved the final manuscript.\u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors are grateful to the Nature Conservation Agency of the Czech Republic for granting the permit to perform environmental sampling within the Zofinsky prales National Natural Reserve. The authors acknowledge BioRender, which was used for figure creation.\u003c/p\u003e\n"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAlleman, A.B., Garcia Costas, A., Mus, F., Peters, J.W., 2022. Rnf and Fix Have Specific Roles during Aerobic Nitrogen Fixation in Azotobacter vinelandii. Appl. Environ. Microbiol. 88, e01049-22. https://doi.org/10.1128/aem.01049-22\u003c/li\u003e\n\u003cli\u003eAramaki, T., Blanc-Mathieu, R., Endo, H., Ohkubo, K., Kanehisa, M., Goto, S., Ogata, H., 2020. KofamKOALA: KEGG Ortholog assignment based on profile HMM and adaptive score threshold. Bioinformatics 36, 2251\u0026ndash;2252. https://doi.org/10.1093/bioinformatics/btz859\u003c/li\u003e\n\u003cli\u003eAroney, S.T.N., Newell, R.J.P., Nissen, J., Camargo, A.P., Tyson, G.W., Woodcroft, B.J., 2024. CoverM: Read coverage calculator for metagenomics. https://doi.org/10.5281/ZENODO.10531253\u003c/li\u003e\n\u003cli\u003eBaldrian, P., L\u0026oacute;pez-Mond\u0026eacute;jar, R., Kohout, P., 2023. Forest microbiome and global change. Nat. Rev. Microbiol. 21, 487\u0026ndash;501. https://doi.org/10.1038/s41579-023-00876-4\u003c/li\u003e\n\u003cli\u003eBertelli, C., Laird, M.R., Williams, K.P., Simon Fraser University Research Computing Group, Lau, B.Y., Hoad, G., Winsor, G.L., Brinkman, F.S., 2017. IslandViewer 4: expanded prediction of genomic islands for larger-scale datasets. Nucleic Acids Res. 45, W30\u0026ndash;W35. https://doi.org/10.1093/nar/gkx343\u003c/li\u003e\n\u003cli\u003eBesemer, J., 2001. GeneMarkS: a self-training method for prediction of gene starts in microbial genomes. Implications for finding sequence motifs in regulatory regions. Nucleic Acids Res. 29, 2607\u0026ndash;2618. https://doi.org/10.1093/nar/29.12.2607\u003c/li\u003e\n\u003cli\u003eBickhart, D.M., Kolmogorov, M., Tseng, E., Portik, D.M., Korobeynikov, A., Tolstoganov, I., Uritskiy, G., Liachko, I., Sullivan, S.T., Shin, S.B., Zorea, A., Andreu, V.P., Panke-Buisse, K., Medema, M.H., Mizrahi, I., Pevzner, P.A., Smith, T.P.L., 2022. Generating lineage-resolved, complete metagenome-assembled genomes from complex microbial communities. Nat. Biotechnol. 40, 711\u0026ndash;719. https://doi.org/10.1038/s41587-021-01130-z\u003c/li\u003e\n\u003cli\u003eBlin, K., Shaw, S., Kloosterman, A.M., Charlop-Powers, Z., van Wezel, G.P., Medema, M.H., Weber, T., 2021. antiSMASH 6.0: improving cluster detection and comparison capabilities. Nucleic Acids Res. 49, W29\u0026ndash;W35. https://doi.org/10.1093/nar/gkab335\u003c/li\u003e\n\u003cli\u003eBoer, W. de, Folman, L.B., Summerbell, R.C., Boddy, L., 2005. Living in a fungal world: impact of fungi on soil bacterial niche development. FEMS Microbiol. Rev. 29, 795\u0026ndash;811. https://doi.org/10.1016/j.femsre.2004.11.005\u003c/li\u003e\n\u003cli\u003eBolger, A.M., Lohse, M., Usadel, B., 2014. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30, 2114\u0026ndash;2120. https://doi.org/10.1093/bioinformatics/btu170\u003c/li\u003e\n\u003cli\u003eBolhuis, H., Severin, I., Confurius-Guns, V., Wollenzien, U.I.A., Stal, L.J., 2010. Horizontal transfer of the nitrogen fixation gene cluster in the cyanobacterium \u003cem\u003eMicrocoleus chthonoplastes\u003c/em\u003e. ISME J. 4, 121\u0026ndash;130. https://doi.org/10.1038/ismej.2009.99\u003c/li\u003e\n\u003cli\u003eBrinkmann, S., Kurz, M., Patras, M.A., Hartwig, C., Marner, M., Leis, B., Billion, A., Kleiner, Y., Bauer, A., Toti, L., P\u0026ouml;verlein, C., Hammann, P.E., Vilcinskas, A., Glaeser, J., Spohn, M., Sch\u0026auml;berle, T.F., 2022. Genomic and Chemical Decryption of the Bacteroidetes Phylum for Its Potential to Biosynthesize Natural Products. Microbiol. Spectr. 10, e02479-21. https://doi.org/10.1128/spectrum.02479-21\u003c/li\u003e\n\u003cli\u003eBrown, C.L., Keenum, I.M., Dai, D., Zhang, L., Vikesland, P.J., Pruden, A., 2021. Critical evaluation of short, long, and hybrid assembly for contextual analysis of antibiotic resistance genes in complex environmental metagenomes. Sci. Rep. 11, 3753. https://doi.org/10.1038/s41598-021-83081-8\u003c/li\u003e\n\u003cli\u003eBuchfink, B., Reuter, K., Drost, H.-G., 2021. Sensitive protein alignments at tree-of-life scale using DIAMOND. Nat. Methods 18, 366\u0026ndash;368. https://doi.org/10.1038/s41592-021-01101-x\u003c/li\u003e\n\u003cli\u003eBurris, R.H., Roberts, G.P., 1993. Biological Nitrogen Fixation. Annu. Rev. Nutr. 13, 317\u0026ndash;335. https://doi.org/10.1146/annurev.nu.13.070193.001533\u003c/li\u003e\n\u003cli\u003eCantalapiedra, C.P., Hern\u0026aacute;ndez-Plaza, A., Letunic, I., Bork, P., Huerta-Cepas, J., 2021. eggNOG-mapper v2: Functional Annotation, Orthology Assignments, and Domain Prediction at the Metagenomic Scale. Mol. Biol. Evol. 38, 5825\u0026ndash;5829. https://doi.org/10.1093/molbev/msab293\u003c/li\u003e\n\u003cli\u003eChaumeil, P.-A., Mussig, A.J., Hugenholtz, P., Parks, D.H., 2019. GTDB-Tk: a toolkit to classify genomes with the Genome Taxonomy Database. Bioinformatics btz848. https://doi.org/10.1093/bioinformatics/btz848\u003c/li\u003e\n\u003cli\u003eCherkasov, N., Ibhadon, A.O., Fitzpatrick, P., 2015. A review of the existing and alternative methods for greener nitrogen fixation. Chem. Eng. Process. Process Intensif. 90, 24\u0026ndash;33. https://doi.org/10.1016/j.cep.2015.02.004\u003c/li\u003e\n\u003cli\u003eChklovski, A., Parks, D.H., Woodcroft, B.J., Tyson, G.W., 2023. CheckM2: a rapid, scalable and accurate tool for assessing microbial genome quality using machine learning. Nat. Methods 20, 1203\u0026ndash;1212. https://doi.org/10.1038/s41592-023-01940-w\u003c/li\u003e\n\u003cli\u003eChoi, B.Y., Lee, S., Kim, J., Park, H., Kim, J.-H., Kim, M., Park, S.-J., Kim, K.-T., Ryu, H., Shim, D., 2022. Comparison of Endophytic and Epiphytic Microbial Communities in Surviving and Dead Korean Fir (Abies koreana) Using Metagenomic Sequencing. Forests 13, 1932. https://doi.org/10.3390/f13111932\u003c/li\u003e\n\u003cli\u003eCriscuolo, A., Gribaldo, S., 2010. BMGE (Block Mapping and Gathering with Entropy): a new software for selection of phylogenetic informative regions from multiple sequence alignments. BMC Evol. Biol. 10, 210. https://doi.org/10.1186/1471-2148-10-210\u003c/li\u003e\n\u003cli\u003eCrits-Christoph, A., Diamond, S., Butterfield, C.N., Thomas, B.C., Banfield, J.F., 2018. Novel soil bacteria possess diverse genes for secondary metabolite biosynthesis. Nature 558, 440\u0026ndash;444. https://doi.org/10.1038/s41586-018-0207-y\u003c/li\u003e\n\u003cli\u003eDavies‐Barnard, T., Friedlingstein, P., 2020. The Global Distribution of Biological Nitrogen Fixation in Terrestrial Natural Ecosystems. Glob. Biogeochem. Cycles 34, e2019GB006387. https://doi.org/10.1029/2019GB006387\u003c/li\u003e\n\u003cli\u003eDos Santos, P.C., Fang, Z., Mason, S.W., Setubal, J.C., Dixon, R., 2012. Distribution of nitrogen fixation and nitrogenase-like sequences amongst microbial genomes. BMC Genomics 13, 162. https://doi.org/10.1186/1471-2164-13-162\u003c/li\u003e\n\u003cli\u003eEdgar, R.C., 2013. UPARSE: highly accurate OTU sequences from microbial amplicon reads. Nat. Methods 10, 996\u0026ndash;998. https://doi.org/10.1038/nmeth.2604\u003c/li\u003e\n\u003cli\u003eEdgar, R.C., 2010. Search and clustering orders of magnitude faster than BLAST. Bioinformatics 26, 2460\u0026ndash;2461. https://doi.org/10.1093/bioinformatics/btq461\u003c/li\u003e\n\u003cli\u003eFeng, X., Cheng, H., Portik, D., Li, H., 2022. Metagenome assembly of high-fidelity long reads with hifiasm-meta. Nat. Methods 19, 671\u0026ndash;674. https://doi.org/10.1038/s41592-022-01478-3\u003c/li\u003e\n\u003cli\u003eFinn, R.D., Clements, J., Eddy, S.R., 2011. HMMER web server: interactive sequence similarity searching. Nucleic Acids Res. 39, W29\u0026ndash;W37. https://doi.org/10.1093/nar/gkr367\u003c/li\u003e\n\u003cli\u003eGalagan, J.E., Henn, M.R., Ma, L.-J., Cuomo, C.A., Birren, B., 2005. Genomics of the fungal kingdom: Insights into eukaryotic biology. Genome Res. 15, 1620\u0026ndash;1631. https://doi.org/10.1101/gr.3767105\u003c/li\u003e\n\u003cli\u003eGrant, J.R., Enns, E., Marinier, E., Mandal, A., Herman, E.K., Chen, C., Graham, M., Van Domselaar, G., Stothard, P., 2023. Proksee: in-depth characterization and visualization of bacterial genomes. Nucleic Acids Res. 51, W484\u0026ndash;W492. https://doi.org/10.1093/nar/gkad326\u003c/li\u003e\n\u003cli\u003eHunt, M., Silva, N.D., Otto, T.D., Parkhill, J., Keane, J.A., Harris, S.R., 2015. Circlator: automated circularization of genome assemblies using long sequencing reads. Genome Biol. 16, 294. https://doi.org/10.1186/s13059-015-0849-0\u003c/li\u003e\n\u003cli\u003eHuson, D.H., Beier, S., Flade, I., G\u0026oacute;rska, A., El-Hadidi, M., Mitra, S., Ruscheweyh, H.-J., Tappu, R., 2016. MEGAN Community Edition - Interactive Exploration and Analysis of Large-Scale Microbiome Sequencing Data. PLOS Comput. Biol. 12, e1004957. https://doi.org/10.1371/journal.pcbi.1004957\u003c/li\u003e\n\u003cli\u003eJiang, F., Li, Q., Wang, S., Shen, T., Wang, H., Wang, A., Xu, D., Yuan, L., Lei, L., Chen, R., Yang, B., Deng, Y., Fan, W., 2023. Recovery of metagenome-assembled microbial genomes from a full-scale biogas plant of food waste by pacific biosciences high-fidelity sequencing. Front. Microbiol. 13, 1095497. https://doi.org/10.3389/fmicb.2022.1095497\u003c/li\u003e\n\u003cli\u003eJohnson, D.C., Unciuleac, M.-C., Dean, D.R., 2006. Controlled Expression and Functional Analysis of Iron-Sulfur Cluster Biosynthetic Components within \u003cem\u003eAzotobacter vinelandii\u003c/em\u003e. J. Bacteriol. 188, 7551\u0026ndash;7561. https://doi.org/10.1128/JB.00596-06\u003c/li\u003e\n\u003cli\u003eKatoh, K., Rozewicki, J., Yamada, K.D., 2019. MAFFT online service: multiple sequence alignment, interactive sequence choice and visualization. Brief. Bioinform. 20, 1160\u0026ndash;1166. https://doi.org/10.1093/bib/bbx108\u003c/li\u003e\n\u003cli\u003eKim, C.Y., Ma, J., Lee, I., 2022. HiFi metagenomic sequencing enables assembly of accurate and complete genomes from human gut microbiota. Nat. Commun. 13, 6367. https://doi.org/10.1038/s41467-022-34149-0\u003c/li\u003e\n\u003cli\u003eKolmogorov, M., Bickhart, D.M., Behsaz, B., Gurevich, A., Rayko, M., Shin, S.B., Kuhn, K., Yuan, J., Polevikov, E., Smith, T.P.L., Pevzner, P.A., 2020. metaFlye: scalable long-read metagenome assembly using repeat graphs. Nat. Methods 17, 1103\u0026ndash;1110. https://doi.org/10.1038/s41592-020-00971-x\u003c/li\u003e\n\u003cli\u003eKuroda, K., Yamamoto, K., Nakai, R., Hirakata, Y., Kubota, K., Nobu, M.K., Narihiro, T., 2022. Symbiosis between \u003cem\u003eCandidatus\u003c/em\u003e Patescibacteria and Archaea Discovered in Wastewater-Treating Bioreactors. mBio 13, e01711-22. https://doi.org/10.1128/mbio.01711-22\u003c/li\u003e\n\u003cli\u003eLangmead, B., Salzberg, S.L., 2012. Fast gapped-read alignment with Bowtie 2. Nat. Methods 9, 357\u0026ndash;359. https://doi.org/10.1038/nmeth.1923\u003c/li\u003e\n\u003cli\u003eLemos, L.N., Mendes, L.W., Baldrian, P., Pylro, V.S., 2021. Genome-Resolved Metagenomics Is Essential for Unlocking the Microbial Black Box of the Soil. Trends Microbiol. 29, 279\u0026ndash;282. https://doi.org/10.1016/j.tim.2021.01.013\u003c/li\u003e\n\u003cli\u003eLi, D., Liu, C.-M., Luo, R., Sadakane, K., Lam, T.-W., 2015. MEGAHIT: an ultra-fast single-node solution for large and complex metagenomics assembly via succinct \u003cem\u003ede Bruijn\u003c/em\u003e graph. Bioinformatics 31, 1674\u0026ndash;1676. https://doi.org/10.1093/bioinformatics/btv033\u003c/li\u003e\n\u003cli\u003eLi, H., 2021. New strategies to improve minimap2 alignment accuracy. Bioinformatics 37, 4572\u0026ndash;4574. https://doi.org/10.1093/bioinformatics/btab705\u003c/li\u003e\n\u003cli\u003eL\u0026oacute;pez-Mond\u0026eacute;jar, R., Brabcov\u0026aacute;, V., \u0026Scaron;tursov\u0026aacute;, M., Davidov\u0026aacute;, A., Jansa, J., Cajthaml, T., Baldrian, P., 2018. Decomposer food web in a deciduous forest shows high share of generalist microorganisms and importance of microbial biomass recycling. ISME J. 12, 1768\u0026ndash;1778. https://doi.org/10.1038/s41396-018-0084-2\u003c/li\u003e\n\u003cli\u003eL\u0026oacute;pez-Mond\u0026eacute;jar, R., Tl\u0026aacute;skal, V., Da Rocha, U.N., Baldrian, P., 2022. Global Distribution of Carbohydrate Utilization Potential in the Prokaryotic Tree of Life. mSystems 7, e00829-22. https://doi.org/10.1128/msystems.00829-22\u003c/li\u003e\n\u003cli\u003eMa, B., Lu, C., Wang, Y., Yu, J., Zhao, K., Xue, R., Ren, H., Lv, X., Pan, R., Zhang, J., Zhu, Y., Xu, J., 2023. A genomic catalogue of soil microbiomes boosts mining of biodiversity and genetic resources. Nat. Commun. 14, 7318. https://doi.org/10.1038/s41467-023-43000-z\u003c/li\u003e\n\u003cli\u003eMarcon, E., Herault, B., 2015. entropart: An R Package to Measure and Partition Diversity. J. Stat. Softw. 68, 1\u0026ndash;26.\u003c/li\u003e\n\u003cli\u003eMarx, V., 2023. Method of the year: long-read sequencing. Nat. Methods 20, 6\u0026ndash;11. https://doi.org/10.1038/s41592-022-01730-w\u003c/li\u003e\n\u003cli\u003eMcMurdie, P.J., Holmes, S., 2013. phyloseq: An R package for reproducible interactive analysis and graphics of microbiome census data. PLoS ONE 8.\u003c/li\u003e\n\u003cli\u003eMu\u0026ntilde;oz-Dorado, J., Marcos-Torres, F.J., Garc\u0026iacute;a-Bravo, E., Moraleda-Mu\u0026ntilde;oz, A., P\u0026eacute;rez, J., 2016. Myxobacteria: Moving, Killing, Feeding, and Surviving Together. Front. Microbiol. 7. https://doi.org/10.3389/fmicb.2016.00781\u003c/li\u003e\n\u003cli\u003eNan, B., Bandaria, J.N., Moghtaderi, A., Sun, I.-H., Yildiz, A., Zusman, D.R., 2013. Flagella stator homologs function as motors for myxobacterial gliding motility by moving in helical trajectories. Proc. Natl. Acad. Sci. 110. https://doi.org/10.1073/pnas.1219982110\u003c/li\u003e\n\u003cli\u003eNavarro-Mu\u0026ntilde;oz, J.C., Selem-Mojica, N., Mullowney, M.W., Kautsar, S.A., Tryon, J.H., Parkinson, E.I., De Los Santos, E.L.C., Yeong, M., Cruz-Morales, P., Abubucker, S., Roeters, A., Lokhorst, W., Fernandez-Guerra, A., Cappelini, L.T.D., Goering, A.W., Thomson, R.J., Metcalf, W.W., Kelleher, N.L., Barona-Gomez, F., Medema, M.H., 2020. A computational framework to explore large-scale biosynthetic diversity. Nat. Chem. Biol. 16, 60\u0026ndash;68. https://doi.org/10.1038/s41589-019-0400-9\u003c/li\u003e\n\u003cli\u003eNayfach, S., Roux, S., Seshadri, R., Udwary, D., Varghese, N., Schulz, F., Wu, D., Paez-Espino, D., Chen, I.-M., Huntemann, M., Palaniappan, K., Ladau, J., Mukherjee, S., Reddy, T.B.K., Nielsen, T., Kirton, E., Faria, J.P., Edirisinghe, J.N., Henry, C.S., Jungbluth, S.P., Chivian, D., Dehal, P., Wood-Charlson, E.M., Arkin, A.P., Tringe, S.G., Visel, A., IMG/M Data Consortium, Abreu, H., Acinas, S.G., Allen, E., Allen, M.A., Alteio, L.V., Andersen, G., Anesio, A.M., Attwood, G., Avila-Maga\u0026ntilde;a, V., Badis, Y., Bailey, J., Baker, B., Baldrian, P., Barton, H.A., Beck, D.A.C., Becraft, E.D., Beller, H.R., Beman, J.M., Bernier-Latmani, R., Berry, T.D., Bertagnolli, A., Bertilsson, S., Bhatnagar, J.M., Bird, J.T., Blanchard, J.L., Blumer-Schuette, S.E., Bohannan, B., Borton, M.A., Brady, A., Brawley, S.H., Brodie, J., Brown, S., Brum, J.R., Brune, A., Bryant, D.A., Buchan, A., Buckley, D.H., Buongiorno, J., Cadillo-Quiroz, H., Caffrey, S.M., Campbell, A.N., Campbell, B., Carr, S., Carroll, J., Cary, S.C., Cates, A.M., Cattolico, R.A., Cavicchioli, R., Chistoserdova, L., Coleman, M.L., Constant, P., Conway, J.M., Mac Cormack, W.P., Crowe, S., Crump, B., Currie, C., Daly, R., DeAngelis, K.M., Denef, V., Denman, S.E., Desta, A., Dionisi, H., Dodsworth, J., Dombrowski, N., Donohue, T., Dopson, M., Driscoll, T., Dunfield, P., Dupont, C.L., Dynarski, K.A., Edgcomb, V., Edwards, E.A., Elshahed, M.S., Figueroa, I., Flood, B., Fortney, N., Fortunato, C.S., Francis, C., Gachon, C.M.M., Garcia, S.L., Gazitua, M.C., Gentry, T., Gerwick, L., Gharechahi, J., Girguis, P., Gladden, J., Gradoville, M., Grasby, S.E., Gravuer, K., Grettenberger, C.L., Gruninger, R.J., Guo, J., Habteselassie, M.Y., Hallam, S.J., Hatzenpichler, R., Hausmann, B., Hazen, T.C., Hedlund, B., Henny, C., Herfort, L., Hernandez, M., Hershey, O.S., Hess, M., Hollister, E.B., Hug, L.A., Hunt, D., Jansson, J., Jarett, J., Kadnikov, V.V., Kelly, C., Kelly, R., Kelly, W., Kerfeld, C.A., Kimbrel, J., Klassen, J.L., Konstantinidis, K.T., Lee, L.L., Li, W.-J., Loder, A.J., Loy, A., Lozada, M., MacGregor, B., Magnabosco, C., Maria Da Silva, A., McKay, R.M., McMahon, K., McSweeney, C.S., Medina, M., Meredith, L., Mizzi, J., Mock, T., Momper, L., Moran, M.A., Morgan-Lang, C., Moser, D., Muyzer, G., Myrold, D., Nash, M., Nesb\u0026oslash;, C.L., Neumann, A.P., Neumann, R.B., Noguera, D., Northen, T., Norton, J., Nowinski, B., N\u0026uuml;sslein, K., O\u0026rsquo;Malley, M.A., Oliveira, R.S., Maia De Oliveira, V., Onstott, T., Osvatic, J., Ouyang, Y., Pachiadaki, M., Parnell, J., Partida-Martinez, L.P., Peay, K.G., Pelletier, D., Peng, X., Pester, M., Pett-Ridge, J., Peura, S., Pjevac, P., Plominsky, A.M., Poehlein, A., Pope, P.B., Ravin, N., Redmond, M.C., Reiss, R., Rich, V., Rinke, C., Rodrigues, J.L.M., Rodriguez-Reillo, W., Rossmassler, K., Sackett, J., Salekdeh, G.H., Saleska, S., Scarborough, M., Schachtman, D., Schadt, C.W., Schrenk, M., Sczyrba, A., Sengupta, A., Setubal, J.C., Shade, A., Sharp, C., Sherman, D.H., Shubenkova, O.V., Sierra-Garcia, I.N., Simister, R., Simon, H., Sj\u0026ouml;ling, S., Slonczewski, J., Correa De Souza, R.S., Spear, J.R., Stegen, J.C., Stepanauskas, R., Stewart, F., Suen, G., Sullivan, M., Sumner, D., Swan, B.K., Swingley, W., Tarn, J., Taylor, G.T., Teeling, H., Tekere, M., Teske, A., Thomas, T., Thrash, C., Tiedje, J., Ting, C.S., Tully, B., Tyson, G., Ulloa, O., Valentine, D.L., Van Goethem, M.W., VanderGheynst, J., Verbeke, T.J., Vollmers, J., Vuillemin, A., Waldo, N.B., Walsh, D.A., Weimer, B.C., Whitman, T., Van Der Wielen, P., Wilkins, M., Williams, T.J., Woodcroft, B., Woolet, J., Wrighton, K., Ye, J., Young, E.B., Youssef, N.H., Yu, F.B., Zemskaya, T.I., Ziels, R., Woyke, T., Mouncey, N.J., Ivanova, N.N., Kyrpides, N.C., Eloe-Fadrosh, E.A., 2021. A genomic catalog of Earth\u0026rsquo;s microbiomes. Nat. Biotechnol. 39, 499\u0026ndash;509. https://doi.org/10.1038/s41587-020-0718-6\u003c/li\u003e\n\u003cli\u003eNguyen, L.-T., Schmidt, H.A., Von Haeseler, A., Minh, B.Q., 2015. IQ-TREE: A Fast and Effective Stochastic Algorithm for Estimating Maximum-Likelihood Phylogenies. Mol. Biol. Evol. 32, 268\u0026ndash;274. https://doi.org/10.1093/molbev/msu300\u003c/li\u003e\n\u003cli\u003eOlm, M.R., Brown, C.T., Brooks, B., Banfield, J.F., 2017. dRep: a tool for fast and accurate genomic comparisons that enables improved genome recovery from metagenomes through de-replication. ISME J. 11, 2864\u0026ndash;2868. https://doi.org/10.1038/ismej.2017.126\u003c/li\u003e\n\u003cli\u003eOrakov, A., Fullam, A., Coelho, L.P., Khedkar, S., Szklarczyk, D., Mende, D.R., Schmidt, T.S.B., Bork, P., 2021. GUNC: detection of chimerism and contamination in prokaryotic genomes. Genome Biol. 22, 178. https://doi.org/10.1186/s13059-021-02393-0\u003c/li\u003e\n\u003cli\u003eParks, D.H., Imelfort, M., Skennerton, C.T., Hugenholtz, P., Tyson, G.W., 2015. CheckM: assessing the quality of microbial genomes recovered from isolates, single cells, and metagenomes. Genome Res. 25, 1043\u0026ndash;1055. https://doi.org/10.1101/gr.186072.114\u003c/li\u003e\n\u003cli\u003eParks, D. H., Rinke, C., Chuvochina, M., Chaumeil, P.A., Woodcroft, B.J., Evans, P.N., et al. (2017). Recovery of nearly 8,000 metagenome-assembled genomes substantially expands the tree of life. Nature Microbiol., 2(11), 1533-1542. https://doi.org/10.1038/s41564-017-0012-7\u003c/li\u003e\n\u003cli\u003ePaul, E.A., Frey, S.D. (Eds.), 2024. Soil microbiology, ecology, and biochemistry, Fifth edition. ed. Elsevier, Amsterdam, Netherlands.\u003c/li\u003e\n\u003cli\u003ePebesma, E., 2018. Simple Features for R: Standardized Support for Spatial Vector Data. R J. 10, 439. https://doi.org/10.32614/RJ-2018-009\u003c/li\u003e\n\u003cli\u003eQuast, C., Pruesse, E., Yilmaz, P., Gerken, J., Schweer, T., Yarza, P., Peplies, J., Gl\u0026ouml;ckner, F.O., 2012. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res. 41, D590\u0026ndash;D596. https://doi.org/10.1093/nar/gks1219\u003c/li\u003e\n\u003cli\u003eRho, M., Tang, H., Ye, Y., 2010. FragGeneScan: predicting genes in short and error-prone reads. Nucleic Acids Res. 38, e191\u0026ndash;e191. https://doi.org/10.1093/nar/gkq747\u003c/li\u003e\n\u003cli\u003eRinne‐Garmston, K.T., Peltoniemi, K., Chen, J., Peltoniemi, M., Fritze, H., M\u0026auml;kip\u0026auml;\u0026auml;, R., 2019. Carbon flux from decomposing wood and its dependency on temperature, wood N \u003csub\u003e2\u003c/sub\u003e fixation rate, moisture and fungal composition in a Norway spruce forest. Glob. Change Biol. 25, 1852\u0026ndash;1867. https://doi.org/10.1111/gcb.14594\u003c/li\u003e\n\u003cli\u003eSagova-Mareckova, M., Cermak, L., Novotna, J., Plhackova, K., Forstova, J., Kopecky, J., 2008. Innovative Methods for Soil DNA Purification Tested in Soils with Widely Differing Characteristics. Appl. Environ. Microbiol. 74, 2902\u0026ndash;2907. https://doi.org/10.1128/AEM.02161-07\u003c/li\u003e\n\u003cli\u003eSaraiva, J.P., Bartholom\u0026auml;us, A., Toscan, R.B., Baldrian, P., Nunes da Rocha, U., 2023. Recovery of 197 eukaryotic bins reveals major challenges for eukaryote genome reconstruction from terrestrial metagenomes. Mol. Ecol. Resour. 23, 1066\u0026ndash;1076. https://doi.org/10.1111/1755-0998.13776\u003c/li\u003e\n\u003cli\u003eSeibold, S., Rammer, W., Hothorn, T., Seidl, R., Ulyshen, M.D., Lorz, J., Cadotte, M.W., Lindenmayer, D.B., Adhikari, Y.P., Arag\u0026oacute;n, R., Bae, S., Baldrian, P., Barimani Varandi, H., Barlow, J., B\u0026auml;ssler, C., Beauch\u0026ecirc;ne, J., Berenguer, E., Bergamin, R.S., Birkemoe, T., Boros, G., Brandl, R., Brustel, H., Burton, P.J., Cakpo-Tossou, Y.T., Castro, J., Cateau, E., Cobb, T.P., Farwig, N., Fern\u0026aacute;ndez, R.D., Firn, J., Gan, K.S., Gonz\u0026aacute;lez, G., Gossner, M.M., Habel, J.C., H\u0026eacute;bert, C., Heibl, C., Heikkala, O., Hemp, A., Hemp, C., Hj\u0026auml;lt\u0026eacute;n, J., Hotes, S., Kouki, J., Lachat, T., Liu, J., Liu, Y., Luo, Y.-H., Macandog, D.M., Martina, P.E., Mukul, S.A., Nachin, B., Nisbet, K., O\u0026rsquo;Halloran, J., Oxbrough, A., Pandey, J.N., Pavl\u0026iacute;ček, T., Pawson, S.M., Rakotondranary, J.S., Ramanamanjato, J.-B., Rossi, L., Schmidl, J., Schulze, M., Seaton, S., Stone, M.J., Stork, N.E., Suran, B., Sverdrup-Thygeson, A., Thorn, S., Thyagarajan, G., Wardlaw, T.J., Weisser, W.W., Yoon, S., Zhang, N., M\u0026uuml;ller, J., 2021. The contribution of insects to global forest deadwood decomposition. Nature 597, 77\u0026ndash;81. https://doi.org/10.1038/s41586-021-03740-8\u003c/li\u003e\n\u003cli\u003eSereika, M., Kirkegaard, R.H., Karst, S.M., Michaelsen, T.Y., S\u0026oslash;rensen, E.A., Wollenberg, R.D., Albertsen, M., 2022. Oxford Nanopore R10.4 long-read sequencing enables the generation of near-finished bacterial genomes from pure cultures and metagenomes without short-read or reference polishing. Nat. Methods 19, 823\u0026ndash;826. https://doi.org/10.1038/s41592-022-01539-7\u003c/li\u003e\n\u003cli\u003eShaffer, M., Borton, M.A., McGivern, B.B., Zayed, A.A., La Rosa, S.L., Solden, L.M., Liu, P., Narrowe, A.B., Rodr\u0026iacute;guez-Ramos, J., Bolduc, B., Gazit\u0026uacute;a, M.C., Daly, R.A., Smith, G.J., Vik, D.R., Pope, P.B., Sullivan, M.B., Roux, S., Wrighton, K.C., 2020. DRAM for distilling microbial metabolism to automate the curation of microbiome function. Nucleic Acids Res. 48, 8883\u0026ndash;8900. https://doi.org/10.1093/nar/gkaa621\u003c/li\u003e\n\u003cli\u003eSharrar, A.M., Crits-Christoph, A., M\u0026eacute;heust, R., Diamond, S., Starr, E.P., Banfield, J.F., 2020. Bacterial Secondary Metabolite Biosynthetic Potential in Soil Varies with Phylum, Depth, and Vegetation Type. mBio 11, e00416-20. https://doi.org/10.1128/mBio.00416-20\u003c/li\u003e\n\u003cli\u003eShen, W., Le, S., Li, Y., \u0026amp; Hu, F., 2016. SeqKit: a cross-platform and ultrafast toolkit for FASTA/Q file manipulation. PloS one, 11(10), e0163962. https://doi.org/10.1371/journal.pone.0163962\u003c/li\u003e\n\u003cli\u003eTerlouw, B.R., Blin, K., Navarro-Mu\u0026ntilde;oz, J.C., Avalon, N.E., Chevrette, M.G., Egbert, S., Lee, S., Meijer, D., Recchia, M.J.J., Reitz, Z.L., van Santen, J.A., Selem-Mojica, N., T\u0026oslash;rring, T., Zaroubi, L., Alanjary, M., Aleti, G., Aguilar, C., Al-Salihi, S.A.A., Augustijn, H.E., Avelar-Rivas, J.A., Avitia-Dom\u0026iacute;nguez, L.A., Barona-G\u0026oacute;mez, F., Bernaldo-Ag\u0026uuml;ero, J., Bielinski, V.A., Biermann, F., Booth, T.J., Carrion Bravo, V.J., Castelo-Branco, R., Chagas, F.O., Cruz-Morales, P., Du, C., Duncan, K.R., Gavriilidou, A., Gayrard, D., Guti\u0026eacute;rrez-Garc\u0026iacute;a, K., Haslinger, K., Helfrich, E.J.N., van der Hooft, J.J.J., Jati, A.P., Kalkreuter, E., Kalyvas, N., Kang, K.B., Kautsar, S., Kim, W., Kunjapur, A.M., Li, Y.-X., Lin, G.-M., Loureiro, C., Louwen, J.J.R., Louwen, N.L.L., Lund, G., Parra, J., Philmus, B., Pourmohsenin, B., Pronk, L.J.U., Rego, A., Rex, D.A.B., Robinson, S., Rosas-Becerra, L.R., Roxborough, E.T., Schorn, M.A., Scobie, D.J., Singh, K.S., Sokolova, N., Tang, X., Udwary, D., Vigneshwari, A., Vind, K., Vromans, S.P.J.M., Waschulin, V., Williams, S.E., Winter, J.M., Witte, T.E., Xie, H., Yang, D., Yu, J., Zdouc, M., Zhong, Z., Collemare, J., Linington, R.G., Weber, T., Medema, M.H., 2023. MIBiG 3.0: a community-driven effort to annotate experimentally validated biosynthetic gene clusters. Nucleic Acids Res. 51, D603\u0026ndash;D610. https://doi.org/10.1093/nar/gkac1049\u003c/li\u003e\n\u003cli\u003eThiery, S., Kaimer, C., 2020. The Predation Strategy of Myxococcus xanthus. Front. Microbiol. 11, 2. https://doi.org/10.3389/fmicb.2020.00002\u003c/li\u003e\n\u003cli\u003eTl\u0026aacute;skal, V., Baldrian, P., 2021. Deadwood-Inhabiting Bacteria Show Adaptations to Changing Carbon and Nitrogen Availability During Decomposition. Front. Microbiol. 12, 685303. https://doi.org/10.3389/fmicb.2021.685303\u003c/li\u003e\n\u003cli\u003eTl\u0026aacute;skal, V., Brabcov\u0026aacute;, V., Větrovsk\u0026yacute;, T., Jomura, M., L\u0026oacute;pez-Mond\u0026eacute;jar, R., Oliveira Monteiro, L.M., Saraiva, J.P., Human, Z.R., Cajthaml, T., Nunes Da Rocha, U., Baldrian, P., 2021a. Complementary Roles of Wood-Inhabiting Fungi and Bacteria Facilitate Deadwood Decomposition. mSystems 6, e01078-20. https://doi.org/10.1128/mSystems.01078-20\u003c/li\u003e\n\u003cli\u003eTl\u0026aacute;skal, V., Brabcov\u0026aacute;, V., Větrovsk\u0026yacute;, T., L\u0026oacute;pez-Mond\u0026eacute;jar, R., Monteiro, L.M.O., Saraiva, J.P., Da Rocha, U.N., Baldrian, P., 2021b. Metagenomes, metatranscriptomes and microbiomes of naturally decomposing deadwood. Sci. Data 8, 198. https://doi.org/10.1038/s41597-021-00987-8\u003c/li\u003e\n\u003cli\u003eTl\u0026aacute;skal, V., Zrůstov\u0026aacute;, P., Vr\u0026scaron;ka, T., Baldrian, P., 2017. Bacteria associated with decomposing dead wood in a natural temperate forest. FEMS Microbiol. Ecol. 93. https://doi.org/10.1093/femsec/fix157\u003c/li\u003e\n\u003cli\u003eTokumoto, U., 2004. Interchangeability and Distinct Properties of Bacterial Fe-S Cluster Assembly Systems: Functional Replacement of the isc and suf Operons in Escherichia coli with the nifSU-Like Operon from Helicobacter pylori. J. Biochem. (Tokyo) 136, 199\u0026ndash;209. https://doi.org/10.1093/jb/mvh104\u003c/li\u003e\n\u003cli\u003eTria, F.D.K., Landan, G., Dagan, T., 2017. Phylogenetic rooting using minimal ancestor deviation. Nat. Ecol. Evol. 1, 0193. https://doi.org/10.1038/s41559-017-0193\u003c/li\u003e\n\u003cli\u003eUliano-Silva, M., Ferreira, J.G.R.N., Krasheninnikova, K., Darwin Tree of Life Consortium, Blaxter, M., Mieszkowska, N., Hall, N., Holland, P., Durbin, R., Richards, T., Kersey, P., Hollingsworth, P., Wilson, W., Twyford, A., Gaya, E., Lawniczak, M., Lewis, O., Broad, G., Martin, F., Hart, M., Barnes, I., Formenti, G., Abueg, L., Torrance, J., Myers, E.W., Durbin, R., Blaxter, M., McCarthy, S.A., 2023. MitoHiFi: a python pipeline for mitochondrial genome assembly from PacBio high fidelity reads. BMC Bioinformatics 24, 288. https://doi.org/10.1186/s12859-023-05385-y\u003c/li\u003e\n\u003cli\u003eUritskiy, G.V., DiRuggiero, J., Taylor, J., 2018. MetaWRAP\u0026mdash;a flexible pipeline for genome-resolved metagenomic data analysis. Microbiome 6, 158. https://doi.org/10.1186/s40168-018-0541-1\u003c/li\u003e\n\u003cli\u003eVětrovsk\u0026yacute;, T., Baldrian, P., Morais, D., 2018. SEED 2: a user-friendly platform for amplicon high-throughput sequencing data analyses. Bioinformatics 34, 2292\u0026ndash;2294. https://doi.org/10.1093/bioinformatics/bty071\u003c/li\u003e\n\u003cli\u003eWang, Y., Gallagher, L.A., Andrade, P.A., Liu, A., Humphreys, I.R., Turkarslan, S., Cutler, K.J., Arrieta-Ortiz, M.L., Li, Y., Radey, M.C., McLean, J.S., Cong, Q., Baker, D., Baliga, N.S., Peterson, S.B., Mougous, J.D., 2023. Genetic manipulation of Patescibacteria provides mechanistic insights into microbial dark matter and the epibiotic lifestyle. Cell 186, 4803-4817.e13. https://doi.org/10.1016/j.cell.2023.08.017\u003c/li\u003e\n\u003cli\u003eWaschulin, V., Borsetto, C., James, R., Newsham, K.K., Donadio, S., Corre, C., Wellington, E., 2022. Biosynthetic potential of uncultured Antarctic soil bacteria revealed through long-read metagenomic sequencing. ISME J. 16, 101\u0026ndash;111. https://doi.org/10.1038/s41396-021-01052-3\u003c/li\u003e\n\u003cli\u003eWest, P.T., Probst, A.J., Grigoriev, I.V., Thomas, B.C., Banfield, J.F., 2018. Genome-reconstruction for eukaryotes from complex natural microbial communities. Genome Res. 28, 569\u0026ndash;580. https://doi.org/10.1101/gr.228429.117\u003c/li\u003e\n\u003cli\u003eWick, R.R., Judd, L.M., Gorrie, C.L., Holt, K.E., 2017. Unicycler: Resolving bacterial genome assemblies from short and long sequencing reads. PLOS Comput. Biol. 13, e1005595. https://doi.org/10.1371/journal.pcbi.1005595\u003c/li\u003e\n\u003cli\u003eWood, D.E., Lu, J., Langmead, B., 2019. Improved metagenomic analysis with Kraken 2. Genome Biol. 20, 257. https://doi.org/10.1186/s13059-019-1891-0\u003c/li\u003e\n\u003cli\u003eWoodcroft, B.J., Singleton, C.M., Boyd, J.A., Evans, P.N., Emerson, J.B., Zayed, A.A.F., Hoelzle, R.D., Lamberton, T.O., McCalley, C.K., Hodgkins, S.B., Wilson, R.M., Purvine, S.O., Nicora, C.D., Li, C., Frolking, S., Chanton, J.P., Crill, P.M., Saleska, S.R., Rich, V.I., Tyson, G.W., 2018. Genome-centric view of carbon processing in thawing permafrost. Nature 560, 49\u0026ndash;54. https://doi.org/10.1038/s41586-018-0338-1\u003c/li\u003e\n\u003cli\u003eZehr, J.P., Capone, D.G., 2020. Changing perspectives in marine nitrogen fixation. Science 368, eaay9514. https://doi.org/10.1126/science.aay9514\u003c/li\u003e\n\u003cli\u003eZhang, H., Yohe, T., Huang, L., Entwistle, S., Wu, P., Yang, Z., Busk, P.K., Xu, Y., Yin, Y., 2018. dbCAN2: a meta server for automated carbohydrate-active enzyme annotation. Nucleic Acids Res. 46, W95\u0026ndash;W101.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Bacteria, biosynthetic gene cluster, carbohydrate-active enzyme, deadwood, decomposition, fungi, metagenome-assembled genome, metagenomics, nitrogen fixation","lastPublishedDoi":"10.21203/rs.3.rs-4181686/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4181686/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eIn forest ecosystems, biological decomposition of deadwood components plays a pivotal role in nutrient cycling and in carbon storage by enriching soils with organic matter. However, deciphering the functional features of deadwood microbiomes is challenging due to their complexity and the limitations of traditional cultivation methods. Our study demonstrates how such limitations can be overcome by describing metagenome composition and function through the analysis of long DNA molecules using the PacBio HiFi platform.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eThe accuracy of PacBio HiFi long-read sequencing emerges as a robust tool for reconstructing microbial genomes in deadwood. It outperformed the routine short-read sequencing and genome sequencing of isolates in terms of the numbers of genomes recovered, their completeness, and representation of their functional potential. We successfully assembled 69 bacterial genomes representing seven out of eight predominant bacterial phyla, including 14 high-quality draft MAGs and 7 nearly finished MAGs. Notably, the genomic exploration extends to Myxococcota, unveiling the unique capacity of Polyangiaceae to degrade cellulose. Patescibacteria contributed to deadwood decomposition processes, actively decomposing hemicellulose and recycling fungal-derived compounds. Furthermore, a novel nitrogen-fixing bacteria within the Steroidobacteriaceae family were identified, displaying interesting genomic adaptations to environmental conditions. The discovered diversity of biosynthetic gene clusters highlights the untapped potential of deadwood microorganisms for novel secondary metabolite production.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eOur study emphasizes new contributors to wood decomposition, especially Polyangiaceae and Patescibacteria for complex and easily decomposable organic matter, respectively. The identification of nitrogen-fixing capabilities within the Steroidobacteraceae family introduces novel perspectives on nitrogen cycling in deadwood. The diverse array of observed biosynthetic gene clusters suggests intricate interactions among deadwood bacteria and promises the discovery of bioactive compounds. Long read sequencing not only advances our understanding of deadwood microbial communities but also demonstrates previously undiscovered functional capacities of the deadwood microbiome. Its application opens promising avenues for future ecological and biotechnological exploration of microbiomes.\u003c/p\u003e","manuscriptTitle":"Pacbio HiFi sequencing sheds light on key bacteria contributing to deadwood decomposition processes","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-04-04 14:48:15","doi":"10.21203/rs.3.rs-4181686/v1","editorialEvents":[{"type":"communityComments","content":1}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"0dd20647-c6cb-418e-a758-2ec620b33da0","owner":[],"postedDate":"April 4th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-05-06T10:35:00+00:00","versionOfRecord":[],"versionCreatedAt":"2024-04-04 14:48:15","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4181686","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4181686","identity":"rs-4181686","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}