Input a fasta of sequence, masks each sequence randomly according to input parameters. Allows identical masking across the input sequences or random masking. See help for more details.
Example:
python random_seqmask.py -T identical -f fake.fa -p 0.01 -m 5 -o fake_mask.fa
usage: random_seqmask.py [-h] [-T [{identical,random}]] -f FASTA_INPUT
[-p PENALTY] [-m MISSING_FACTOR] -o OUT_FILE
optional arguments:
-h, --help show this help message and exit
-T [{identical,random}], --mask_type [{identical,random}]
Apply identical masking across sequences, requires all
input sequence to be the same length, OR applies
random masking to each sequence, using the same
masking parameters.
-f FASTA_INPUT, --fasta_input FASTA_INPUT
Input is alignment file in fasta format.
-p PENALTY, --penalty PENALTY
Positive real. All else equal, determines how long
islands of missing bases will be. Smaller values means
shorter islands.
-m MISSING_FACTOR, --missing_factor MISSING_FACTOR
Positive real. All else equal, determines how frequent
missing bases will be. Smaller values result in more
missingness.
-o OUT_FILE, --out_file OUT_FILE
The file to write output to.
The repo includes two other helpful scripts: fake_seq.py and count_missing.py, which create a really simple fasta sequence to test the masking out on, and summarize the resulting run lengths of missing and non-missing bases, respectively.
Usage:
python fake_seq.py #modify size and number of sequence in script
#fake_seq.py write fake_mask.fa
python count_missing.py fake_mask.fa