Kmers module
KMERS_GWAS module decompose reads into kmers and create a binary table of presence/absence of kmers. This table can be filter to use only most informative kmers into the populations. PLINK format outfile are obtained in this module.
PARAMS:
KMERS_MODULE:
KMER_SIZE : 31
MAC : 2
P : 0.2
MAF : 0.05
B : 1000000 # nb kmers in each bed file
SPLIT_LIST_SIZE : 100000
MIN_LIST_SIZE : 50000
SAMPLES_FILE: "samples.txt"
Type |
Description |
Default Value |
|---|---|---|
KMER_SIZE |
Length of kmers (should be between 15-31) |
31 |
MAC |
Minor allele count (min allowed appearance of a kmer) |
2 |
P |
Minimum percent of appearance in each strand form |
0.2 |
MAF |
Minimum allele frequency |
0.05 |
B |
Number of kmers in each bed file |
1000000 |
SPLIT_LIST_SIZE |
Number of kmers by bed file |
100000 |
MIN_LIST_SIZE |
Minimal number of kmers allowed in the smaller bed file after splitting |
50000 |
Note
if you active KMERS_MODULE and no active PCADAPT and LFMM, iKISS calculate occurrences of presence and absence of kmers into the populations using the whole of kmers and groups given by user into the SAMPLES_FILE. If MAPPING and INTERSECT are also activated you can check occurrences found into chosen features (genes). Check the ‘go away in global mode’ section.