ISonets/taxo_class_snakemake

Snakemake file and R scripts for producing taxonomic classification of metagenomic RNA reads

taxo_class_snakemake

Snakemake file and R scripts for producing taxonomic classification of metagenomic RNA reads made in Laboratory of Bioinformatics at Research Institute for Systems Biology and Medicine (RISBM).

Prerequisites:

• Snakemake (https://snakemake.github.io/);

• Kraken2 >= 2.1.2 (https://github.com/DerrickWood/kraken2);

  Note: kraken2 needs to be installed into another conda env or installed from source due to dependecy conflicts between snakemake and kraken2.

  Note: use pluspf database for kraken2 (https://benlangmead.github.io/aws-indexes/k2).

• CAT/BAT and its local database (https://github.com/dutilh/CAT);

• Hisat2 (http://daehwankimlab.github.io/hisat2/);

• samtools (https://github.com/samtools/samtools);

• bedtools (https://github.com/arq5x/bedtools2);

• SPAdes (https://github.com/ablab/spades);

• BLAST and its local nt database (https://blast.ncbi.nlm.nih.gov/Blast.cgi);

• R and its libraries:

  • taxonomizr (https://github.com/sherrillmix/taxonomizr),
  • dplyr (https://dplyr.tidyverse.org/),
  • readxl (https://readxl.tidyverse.org/),
  • reshape (https://cran.r-project.org/web/packages/reshape/index.html);

  Note: for taxonomizr library you need local database, but for our purposes accession ids do not needed, to achieve this, use getAccessions=FALSE

Usage:

1. Install all tools and verify if it's working;
2. Open Snakemake file and provide paths to FASTQ files;
3. Launch it and make some tea while it's doing science;
4. Analyse results provided for you in human-readable form:)

Note: you might also need to modify paths in R scripts to accomodate your data and folder structure, because they are hardcoded now.

Authors

Ignat Sonets

Alexander Manolov, PhD