I've been testing out KMC with the following dummy sequence <div class="snippet-cl

kmer counts incorrect? about kmc HOT 2 CLOSED

refresh-bio commented on August 24, 2024

kmer counts incorrect?

from kmc.

Comments (2)

marekkokot commented on August 24, 2024

Hi,

Thanks for using KMC!

In most applications, k-mer and its reverse complement should not be distinguished.
Simply speaking for each k-mer its reverse complement is calculated.
Canonical k-mer is a lexicographically smaller of two (k-mer and its reverse complement). KMC in default mode counts only canonical k-mers.

If you need k-mers as they are, there is an -b option. Then for you sequence the output is:

Stage 1: 100%
1st stage: 0.0552348s
2nd stage: 0.0376972s
Total    : 0.092932s
Tmp size : 0MB

Stats:
   No. of k-mers below min. threshold :            0
   No. of k-mers above max. threshold :            0
   No. of unique k-mers               :           34
   No. of unique counted k-mers       :           34
   Total no. of k-mers                :           51
   Total no. of sequences             :            1
   Total no. of super-k-mers          :            0

There are in fact 34 k-mers:

AAA	3
AAT	4
AGC	1
ATA	2
ATC	2
ATG	2
CAA	1
CCA	1
CCC	1
CCT	1
CGA	1
CGC	4
CTA	2
CTC	1
CTG	1
GAT	1
GCC	1
GCG	2
GCT	2
GGG	1
GGT	1
GTC	1
GTT	1
TAA	2
TAG	1
TAT	1
TCC	1
TCG	3
TCT	1
TGC	1
TGG	1
TGT	1
TTC	1
TTT	1

If you agree and have no questions please close this issue, in the opposite case let me know what is not clear.

from kmc.