Er’s instruction (Illumina). Briefly, DNA fragmentation was carried out by Covaris S2 (Covaris, 01801-1721). The flow cell was sequenced by BGI (Shenzhen, China) using Illumina HiSeq2000 technology generating 26100 bp paired-end reads. The base-calling pipeline (version Illumina Pipeline-0.3) was used to process the raw fluorescence images and call sequences.Metagenome Assembly and Coverage StatisticsThe paired-end sequences were quality-checked by discarding any read containing ambiguous base of letter N and then trimming off the sequencing adaptors to get reads of 100 bp in length. Reads after Licochalcone A quality control (11,930,760 reads, 1.2 Gb) had been submitted to NCBI SRA database under accession number of SRA057365. Reads after quality check were get 16960-16-0 assembled using velvet (version 0.7.62) [17] by options: k = 31, cov_cutoff = 3, and exp_cov = 14. Only contigs longer than 1 kb were used in the following analysis [11]. The read coverage of each ORF/contig was quantified via aligning the paired-end reads to ORFs/contigs by Bowtie (0.12.7) [22]. For each read, only the best alignment was kept, allowing up to 2 mismatches over the entire length (Bowtie options: -k 1 -f 2 est airtries 200 hunkmbs 50000). Gap was not allowed during the alignment. And the statistics of alignment coverage was carried out by the SAMtools package [23]. The assembled contigs longer than 1 kb were subject to gene prediction using online MetaGeneMark [24] under Kingdom of mixture of Bacteria and Archaea with 6-LBA model (Di-codon frequencies fit by logistic regressions on GC content). Totally 31,499 open reading frames were defined by the MetaGeneMark tool.Validation of Metagenome AssemblyTo verify the accuracy of assembly, a random subset of 10 assembled candidate genes were selected to design specific primer 11967625 sets. PCR (polymerase chain reaction) were conducted using the designed primers, respectively, with the same extracted DNA as the template. The PCR condition was as follow: an initial 5 min denaturation at 94uC, 35 cycles of 1 min at 94uC and 3.5 min at 68uC, and a final 5 min extension at 68uC. Each PCR volume contained 50 ng DNA, 200 nM of each primer (Integrated DNA Technologies, US), 25 mL 2 6 Premix Ex Taq (TaKaRa, China), and ddH2O up 
to 50 mL. If the determined sizes of the PCR products matched the sizes of the assembled genes, the purified product (PureLinkTM PCR Purification Kit, Invitrogen, US) was sent out for Sanger sequencing (GRC, The University of Hong Kong). Genes longer than 400 AA were sequenced from both forward and reverse ends.Materials and Methods Thermophilic Anaerobic Cellulolytic SludgeAnaerobic digestion sludge (collected from Shek Wu Hui wastewater treatment plant Hong Kong SAR, China) was used to enrich cellulolytic methanogenic consortia in a sequential batch reactor (SBR) for two years at 55uC and pH .6.0 by feeding microcrystalline cellulose as substrate and glucose as co-substrate at COD ratio of 10:1. Our previous study showed that theCarbohydrate-active Gene 
PredictionNext, amino acid sequences of the predicted ORFs were screened against PfamA database version 26.0 [25] by Pfam_scan (E-value cutoff of 1E-4) [26] for particular glycoside hydrolase (GH) families and carbohydrate binding module (CBM) asMetagenomic Mining of Cellulolytic GenesFigure 5. Comparison of predicted carbohydrate-active genes (top chart) and carbohydrate-binding modules (bottom chart) in three cellulosic materials fed metagenomes: rumen microbiome [11], termi.Er’s instruction (Illumina). Briefly, DNA fragmentation was carried out by Covaris S2 (Covaris, 01801-1721). The flow cell was sequenced by BGI (Shenzhen, China) using Illumina HiSeq2000 technology generating 26100 bp paired-end reads. The base-calling pipeline (version Illumina Pipeline-0.3) was used to process the raw fluorescence images and call sequences.Metagenome Assembly and Coverage StatisticsThe paired-end sequences were quality-checked by discarding any read containing ambiguous base of letter N and then trimming off the sequencing adaptors to get reads of 100 bp in length. Reads after quality control (11,930,760 reads, 1.2 Gb) had been submitted to NCBI SRA database under accession number of SRA057365. Reads after quality check were assembled using velvet (version 0.7.62) [17] by options: k = 31, cov_cutoff = 3, and exp_cov = 14. Only contigs longer than 1 kb were used in the following analysis [11]. The read coverage of each ORF/contig was quantified via aligning the paired-end reads to ORFs/contigs by Bowtie (0.12.7) [22]. For each read, only the best alignment was kept, allowing up to 2 mismatches over the entire length (Bowtie options: -k 1 -f 2 est airtries 200 hunkmbs 50000). Gap was not allowed during the alignment. And the statistics of alignment coverage was carried out by the SAMtools package [23]. The assembled contigs longer than 1 kb were subject to gene prediction using online MetaGeneMark [24] under Kingdom of mixture of Bacteria and Archaea with 6-LBA model (Di-codon frequencies fit by logistic regressions on GC content). Totally 31,499 open reading frames were defined by the MetaGeneMark tool.Validation of Metagenome AssemblyTo verify the accuracy of assembly, a random subset of 10 assembled candidate genes were selected to design specific primer 11967625 sets. PCR (polymerase chain reaction) were conducted using the designed primers, respectively, with the same extracted DNA as the template. The PCR condition was as follow: an initial 5 min denaturation at 94uC, 35 cycles of 1 min at 94uC and 3.5 min at 68uC, and a final 5 min extension at 68uC. Each PCR volume contained 50 ng DNA, 200 nM of each primer (Integrated DNA Technologies, US), 25 mL 2 6 Premix Ex Taq (TaKaRa, China), and ddH2O up to 50 mL. If the determined sizes of the PCR products matched the sizes of the assembled genes, the purified product (PureLinkTM PCR Purification Kit, Invitrogen, US) was sent out for Sanger sequencing (GRC, The University of Hong Kong). Genes longer than 400 AA were sequenced from both forward and reverse ends.Materials and Methods Thermophilic Anaerobic Cellulolytic SludgeAnaerobic digestion sludge (collected from Shek Wu Hui wastewater treatment plant Hong Kong SAR, China) was used to enrich cellulolytic methanogenic consortia in a sequential batch reactor (SBR) for two years at 55uC and pH .6.0 by feeding microcrystalline cellulose as substrate and glucose as co-substrate at COD ratio of 10:1. Our previous study showed that theCarbohydrate-active Gene PredictionNext, amino acid sequences of the predicted ORFs were screened against PfamA database version 26.0 [25] by Pfam_scan (E-value cutoff of 1E-4) [26] for particular glycoside hydrolase (GH) families and carbohydrate binding module (CBM) asMetagenomic Mining of Cellulolytic GenesFigure 5. Comparison of predicted carbohydrate-active genes (top chart) and carbohydrate-binding modules (bottom chart) in three cellulosic materials fed metagenomes: rumen microbiome [11], termi.
