Example bowtie2

1. Make a clean working directory and download and extract the SLIMM executable inside it. For example (on a Mac),

    mkdir ~/slimm_tutorial
    cd ~/slimm_tutorial
    wget https://github.com/seqan/slimm/releases/download/v0.2/slimm-0.2-Mac-x86_64.zip
    unzip slimm-0.2-Mac-x86_64.zip
   

2. Download and extract the 5K version bowtie2 index (AB_5K_indexed_ref_genomes_bowtie2.zip) from here. This bowtie2 index contains reference genome database of 4538 complete bacterial and archaeal genomes that spans around 2500 different species. For example,

    wget http://ftp.mi.fu-berlin.de/pub/dadi/slimm/AB_5K_indexed_ref_genomes_bowtie2.zip
    unzip AB_5K_indexed_ref_genomes_bowtie2.zip	
   

3. Download and extract the 5K version of SLIMM database (slimmDB_5K.zip) from here. For example,

    wget http://ftp.mi.fu-berlin.de/pub/dadi/slimm/slimmDB_5K.zip
    unzip slimmDB_5K.zip
   

4. Create two new directories with the name alignment_files slimm_reports and place your metagenomic sequencing reads under a directory named mg_reads .

Now you should have the following folder/directory structure in your working directory:

    Working Directory
      │
      ├── bin
      ├    ├── slimm (slimm binary executable)
      │
      ├── AB_5K_indexed_ref_genomes_bowtie2 (indexed reference genomes)
      ├    ├── AB_5K.1.bt2l
      ├    ├── AB_5K.2.bt2l  
      ├    ├── AB_5K.3.bt2l
      ├    ├── ...
      │
      ├── slimmDB_5K (SLIMM taxonomic database)
      │
      ├── alignment_files (alignment files will be stored here)
      │
      ├── slimm_reports (slimm taxonomic reports will be stored here)
      │
      ├── mg_reads (metagenomic sequencing reads) 
      ├    ├── SRR1748536_1.fastq
      ├    ├── SRR1748536_2.fastq

1. Use bowtie2 to map the metagenomic reads against reference genomes and produce alignment files.

    bowtie2 -x .AB_5K_indexed_ref_genomes_bowtie2/AB_5K \
            -1 ./mg_reads/SRR1748536_1.fastq  \
            -2 ./mg_reads/SRR1748536_1.fastq  \
            -q --no-unal --mm -p 10 -k 60  \
            -S ./alignment_files/SRR1748536.sam
   

[recomended] Alternatively we can directly pipe the output to samtools to save it as BAM format. This is even faster than just writing into SAM files since it avoids a lot of IO operations. And SLIMM performs sligtly better on BAM format.

    bowtie2 -x .AB_5K_indexed_ref_genomes_bowtie2/AB_5K \
            -1 ./mg_reads/SRR1748536_1.fastq \
            -2 ./mg_reads/SRR1748536_1.fastq \
            -q --no-unal --mm -p 10 -k 60 \
            2>./alignment_files/SRR1748536.txt | samtools view -bS - > ./alignment_files/SRR1748536.bam;

1. Run SLIMM on the output of the read mapper (SAM/BAM files)

    slimm -m ./slimmDB_5K  -o slimm_reports/ ./alignment_files/SRR1748536.bam
   

you will find multiple reports under the directory slimm_reports for each level of taxonomy (species-superkingdom) named as SRR1748536_[rank]_reported.tsv contains the rank level taxonomic profile of the sample that SLIMM reported.

You can also tell SLIMM to report only a single rank using -r parameter. For example,

    slimm -m ./slimmDB_5K  -o slimm_reports/ -r species ./alignment_files/SRR1748536.bam

Would generate species level report only.

(see slimm --help for more details)