Generation of simulated read sets

Two sets of reads are needed:

• Endogenous reads mimic ancient DNA reads which shall be mapped to a reference
• Exogenous reads mimic contaminant DNA which should not be mapped to the reference

Generating endogenous reads

In this tutorial, endogenous reads are generated from the sample genome volpertinger.fasta, which is stored in the input/genome folder of this manual. The script uniform is used to sample reads uniformly from the reference:

scripts/uniform input/genome/volpertinger.fasta \
    --seed 1234 \
    --name volpertinger_ \
    --output-fastq data/2/volpertinger.coord data/2/volpertinger.fastq \
    25 20 5

The call is explained in the following:

The output files are volpertinger.fastq and volpertinger.coord. They contain the generated reads and the positions on the reference from which they were extracted, respectively.

The parameter --seed is used to initialize the random number generator to a defined state. Calling uniform with same seed value and same input yields the same set of reads. If you don't specify --seed, you will generate different reads each time you call this script.

The parameter --name specifies a prefix for the names the generated reads get, to identify them in the SAM file once they were mapped.

Generating exogenous reads

In this tutorial, contaminant reads are generated from the genome retli_tr.fasta, which is a part of the Rhizobium etli genome.

The process of read generation is very similar to the previous section:

scripts/uniform input/retli/retli_tr.fasta \
    --seed 1235 \
    --name retli_ \
    --output-fastq data/2/retli.coord data/2/retli.fastq \
    50 20 5

Read quality scores

The generated reads have only FFFF..... as base quality scores, if you want more control over the read generation process, see the section "Advanced read generation".