References: Bacteria - AndreaRP/METAGENOMICS GitHub Wiki

Bacteria

Create a directory named BACTERIA_GENOME_REFERENCES/ and go inside it.

[16S] (http://greengenes.secondgenome.com/downloads/database/13_5)

  1. Create a new directory BACTERIA_GENOME_REFERENCES/16S/ and go inside it.

  2. Download the multifasta:

`curl 'https://s3.amazonaws.com/gg_sg_web/gg_13_5.fasta.gz' > bact_16S.fasta.gz`

  1. Decompress:

    tar xvzf bact_16S.fasta.gz

  2. Create bowtie index files.

    bowtie2-build --offrate 3 REFERENCES/BACTERIA_GENOME_REFERENCE/16S/bact_16S.fasta REFERENCES/BACTERIA_GENOME_REFERENCE/16S/bwt2/16S
    The offrate option will make the index building a little slower, but will provide faster mapping later.

  3. Get length of each 16S sequence

    samtools faidx BACTERIA_GENOME_REFERENCE/16/bact_16S.fasta
    (generates index: bact_16S.fasta.fai)

    cut -f1-2 BACTERIA_GENOME_REFERENCE/16S/bact_16S.fasta.fai > BACTERIA_GENOME_REFERENCE/16S/16S_seq_length.txt

[Whole Genome] (ftp://ftp.ncbi.nlm.nih.gov/genomes/refseq/bacteria/)

  1. Create a directory BACTERIA_GENOME_REFERENCES/WG/ and go inside it.

  2. Download the assembly summary file, with a list of all the bacterias included.

    curl 'ftp://ftp.ncbi.nlm.nih.gov/genomes/refseq/bacteria/assembly_summary.txt' > assembly_summary_bacteria.txt

  3. Get a list of complete and latest sequences.

    awk -F "\t" '$12=="Complete Genome" && $11=="latest"{print $1,$20}' assembly_summary_bacteria.txt > ftpdirpaths_complete_latest.txt

  4. Download the fasta sequences.

    cat ftpdirpaths_complete_latest.txt | while read line; do dir=`echo $line | cut -d " " -f 1 `;mkdir -p $dir;url=`echo $line | cut -d " " -f 2`;echo "wget -P $dir $url/*" ; done >> wget_all_bacteria_complete_latest.sh (creates download script)

    bash wget_all_bacteria_complete_latest.sh (executes download script)

    WARNING! This step will create a folder per genome and download several files into it, including nucleotide and protein sequences. Before executing this command please make sure you have sufficient storage space. We advise you create a new directory and execute this command from inside it, that will keep your working space well organized. In this example we have created the folder all_genomes for the download.

  5. Create the multifasta file with the genomic data.

    find BACTERIA_GENOME_REFERENCE/WG/all_genomes/ ! -name '*from_genomic.fna.gz' -name '*_genomic.fna.gz' -exec zcat {} \; > BACTERIA_GENOME_REFERENCE/WG/bacteria_genome_all.fna

  6. Create multifasta file with proteomic data.

    find BACTERIA_GENOME_REFERENCE/WG/all_genomes/ -name '*_protein.faa.gz' -exec zcat {} \; > BACTERIA_GENOME_REFERENCE/WG/bacteria_protein_all.faa

  7. Get length of each sequence

    samtools faidx BACTERIA_GENOME_REFERENCE/WG/bacteria_genome_all.fna (generates index: bacteria_genome_all.fna.fai)

    cut -f1-2 BACTERIA_GENOME_REFERENCE/WG/bacteria_genome_all.fna.fai > BACTERIA_GENOME_REFERENCE/WG/genome_length.txt

  8. Create Bowtie2 genome index

`bowtie2-build REFERENCES/BACTERIA_GENOME_REFERENCE/WG/bacteria_genome_all.fna REFERENCES/BACTERIA_GENOME_REFERENCE/WG/bwt2/WG`
  1. Create BLAST files
`makeblastdb -in REFERENCES/BACTERIA_GENOME_REFERENCE/WG/bacteria_genome_all.fna -parse_seqids -dbtype nucl -out /REFERENCES/BACTERIA_GENOME_REFERENCE/BLAST/blastn/BACTERIA_blastn`