You can find here the bash script used in "Revealing microbial species diversity using sequence capture by hybridization" to reconstruct near full-length 16S rDNA sequences from sequencing data gained by gene capture.

This script takes as input paired end reads (fastq file). We considered that sequence adaptater were removed by the sequencing platform.

Tools necessary to run this script are listed below.

• For quality trimming: Prinseq-lite
• To filter 16S reads: SortMeRNA v2.1
• To reconstruct full-length SSU rRNA sequences: EMIRGE v0.60
• Classify the full-length reconstructed SSU sequences: the Python script assign_taxonomy.py of QIIME v1.9.1 and the SILVA119 QIIME compatible database.

Please install these tools in your $PATH according to their author's recommendations.

SortMeRNA and EMIRGE were fed with the SILVA SSU 119 database which need to be indexed separately for each tool.

SortMeRNA provided rRNA databases in the 'sortmerna- 2.1/rRNA databases' folder. To index the 16S rRNA database use the following line.

sortmerna-2.1-linux-64/indexdb_rna --ref \
./rRNA_databases/silva-bac-16s-id90.fasta,./index/silva-bac-16s-db:\
./rRNA_databases/silva-arc-16s-id95.fasta,./index/silva-arc-16s-db:\
./rRNA_databases/silva-euk-18s-id95.fasta,./index/silva-euk-18s-db

EMIRGE provided a specific script to download and index the current version of the Silva SSU rRNA database (SILVA119 in this work). Run the following command.

python emirge_makedb.py

Run the following command.

bash CaptOTU_16S_reconstruction.sh