PyGargammel is a reimplementation the ancient dna damage simulator GARGAMMEL focused on better exposing the model parameters for simulation. To this end, it is quite barebones, and only implements the Brigg's model of ADNA damage. Unlike GARGAMMEL, PyGargammel does not do contamination of samples, instead choosing to adopt a unix philosophy of doing just one thing. In this case, it is implementing and exposing the parameters of the Brigg's Model.
An example command for PyGargammel is
pygargammel --fasta align.fasta --nick-freq 0.24 --overhang-parameter 0.36 --double-strand-deamination 0.0097
--single-strand-deamination 0.22 --output damage.fasta --log pygargammel.log
Each of the options specified in the example command are required. Here is a quick and informal explanation of the options:
--fasta: The source sequences to damage. If more than one sequence is present PyGargammel will attempt to damage each sequence individually.--nick-freq: Probability of nicks per base. Has the effect of controlling the length and number of fragments. Higher numbers mean more and shorter fragments. Accepts values between 0.0 and 1.0, where 0.0 is no chance of nicks, and 1.0 is a nick on every base.--overhang-parameter: Parameter to the geometric distribution determining the length of the overhangs. Somewhat counter intuitively, higher numbers mean shorter overhangs. Accepts values in the range(0.0, 1.0]. Value close to zero give very long overhangs, and 1.0 will give zero length overhangs.--double-strand-deamination: Rate at deamination will occur in double stranded regions. Accepts values between 0.0 and 1.0, with 0.0 being no deamination, and 1.0 being total deamination.--single-strand-deamination: Rate at deamination will occur in single strand regions. Accepts values between 0.0 and 1.0, with 0.0 being no deamination, and 1.0 being total deamination.--reference: Ensure that generated fragments have at least one informative character in common with all sequence in the alignment.