322    ◾    Bioinformatics

Before we perform the taxonomic binning, we need to generate sequence depth from the

sorted BAM files produced by mapping the metagenomic FASTQ reads to the de novo

assemblies. For this purpose, we will also need the tables produced by “get_count_table.

py” script above as an input with the sorted BAM file for the “jgi_summarize_bam_con-

tig_depths” function to produce a file of five columns: contig name, contig length, total

average depth, mean depth, and variance.

mkdir stats_metabat

jgi_summarize_bam_contig_depths \

--outputDepth stats_metabat/healthy_depth.txt \

sam_assemblies/ERR1823587_healthy.bam.sorted

jgi_summarize_bam_contig_depths \

--outputDepth stats_metabat/moderate_depth.txt \

sam_assemblies/ERR1823601_moderate.bam.sorted

jgi_summarize_bam_contig_depths \

--outputDepth stats_metabat/severe_depth.txt \

sam_assemblies/ERR1823608_severe.bam.sorted

Then, we can perform binning on the contigs.fasta produced by de novo assembly above.

We can copy these files in a new directory “binning” with new names.

mkdir binning

cp metag_healthy/contigs.fasta binning/healthy_contigs.fasta

cp metag_moderate/contigs.fasta binning/moderate_contigs.fasta

cp metag_severe/contigs.fasta binning/severe_contigs.fasta

The next step is to separate the contigs in the contigs files into bins; each bin represents a

species. The bins of the three samples are saved in different subdirectories inside “binning”

directory.

mkdir binning/healthy

metabat2 -i binning/healthy_contigs.fasta \

-a stats_metabat/healthy_depth.txt \

-o binning/healthy/healthy \

-t 4 -v --seed 123

mkdir binning/moderate

metabat2 -i binning/moderate_contigs.fasta \

-a stats_metabat/moderate_depth.txt \

-o binning/moderate/moderate \

-v --seed 123

mkdir binning/severe

metabat2 -i binning/severe_contigs.fasta \

-a stats_metabat/severe_depth.txt \

-o binning/severe/severe \

-v --seed 123