Leif™

The Leif™ Microbiome Analyzer is a set of bioinformatics tools designed to identify microbes in clinical specimens without the use of consensus primers. Performing high-throughput sequencing without consensus primers results in an unbiased survey of the DNA in a sample (Lipkin 2010). Such a survey is capable of identifying microbes from entire tree of life, which is a major advantage as compared to primer based assays such as IBIS or PCR. See the Workflow page or Laurence et al 2014 for a detailed description. 

Figure 1: Leif workflow overview.

The Leif Microbiome Analyzer is currently undergoing beta testing. Beta licenses are free. The latest version can be downloaded here.

Minimum system requirements
  • Microsoft Windows XP (or later)
  • 1 TB of free disk space
  • 2 GB of RAM
Setup steps
  1. Download Leif zip file.
  2. Extract zip folder on a disk that has at least 1TB of free space.
  3. Setup Leif libraries by running one of these scripts:
    1. library_setup.bat full: this downloads NCBI taxonomy data and the four largest NCBI BLAST databases (~1000GB), and builds indexes based on these files (~50GB). This takes 30 hours to run (see Note 1).
    2. library_setup.bat: this downloads NCBI taxonomy data and the smallest NCBI BLAST database (~80 GB), and builds indexes based on these files (~50GB). This takes 3 hours to run (see Note 1).
  4. If you have never used Leif before, you should run a tutorial (these require a 64-bit operating system, see Note 3).
    1. 1_1000gp_run.bat: downloads and aligns a single 1000 Genomes Project run. This takes 10 hours to run (see Note 1, 2).
    2. 6_1000gp_runs.bat: downloads and aligns six 1000 Genomes Project runs. This takes 18 hours to run (see Note 1, 2).
Note 1: Running on an Amazon AWS hi1.4xlarge instance (Intel Xeon E5620 2.40GHz). 

Note 2: Most of the CPU time in the tutorials is spent running the fastq-dump program which converts runs from the proprietary NCBI SRA format to FASTQ format.

Note 3: 1000 Genomes Project reads are stored in SRA format, and must be converted to FASTQ format by an NCBI SRA tool called fastq-dump. This tool requires the win64 environment.

Complete documentation