Fig. 1

Statistics of metagenomic sequencing of 13 faecal samples with the PacBio SMRT system and short-read sequencers. a To show the length distribution of the contigs of long and short reads, we selected three samples (apr34, apr38, and FAKO02) that had similar sequence amounts in both PacBio long-read and short-read sequencing (see the “Results” section). The y-axis shows the Nxx contig length, an indicator of measuring the quality of genome assembly such that xx% of all bases in the assembled contigs of the three selected samples are found in contigs of the Nxx contig length or more, while the x-axis shows the value of xx, which measures coverage of bases by contigs. b Sequence similarity between PacBio and short-read contigs. The y-axis shows the sequence similarity of the PacBio contigs with the reciprocally best-matched short-read contigs, and the plots show the average value for every five units of read depth of the PacBio contigs on the x-axis. PacBio and short-read contigs of the 12 samples were aligned using NUCmer with a > 95% identity and a > 95% length coverage. c Genus-level microbial compositions estimated from the PacBio and MiSeq data of the 13 samples. Taxonomic assignment and quantification of microbial abundance from the PacBio and MiSeq data were described in the “Methods” section. d Pearson’s correlation coefficients (PCCs) between the microbial compositions estimated from PacBio and MiSeq data. PCCs (left) between the same samples, excluding the biological replicates (ES1-2 and ES9-1), and PCCs (right) between different samples are shown. The boxes represent the inter-quartile range (IQR), and the lines inside represent the median. The whiskers show the lowest and highest values within 1.5 times the IQR. Asterisks represent P < 0.01 (Wilcoxon rank-sum test)