Question in terminology

[joskee]

Dear all,

I received a file with some explenation about a sequence, however there is something I dont understand.


"A K-mer refers to an artificial sequence division of K nucleotides. A raw sequencing read with L bp contains (L-K+1) K-mers if the length of each K-mer is K bp. The frequency of each K-mer can be calculated from the raw genome sequencing reads"

I know what a K-mer is, but where do they get the L bp from?
What is this L bp?

[JackieBadger]

Length in bases Id guess.. Makes sense

[joskee]

Length in bases Id guess.. Makes sense

Yes I know its a length of bases, but what does it represent?

What is the L bp in relationship to the K-mer?

Is the K-mer a piece of the L bp ?

[mastal]

Question in terminology

L is the length of the read in base pairs, and yes, the k-mer is a piece of the read.

For example, if you had a read that was 50 bp long, and wanted to use k-mers of length 47, then you would get 50-47+1=4 kmers from your read.

The first kmer would be bases 1-47 of the read, the second kmer would be bases 2-48 of the read, the third kmer bases 3-49 and the fourth one bases 4-50, assuming you were also using a step size of 1.

Are you doing de novo assembly?

[joskee]

L is the length of the read in base pairs, and yes, the k-mer is a piece of the read.

For example, if you had a read that was 50 bp long, and wanted to use k-mers of length 47, then you would get 50-47+1=4 kmers from your read.

The first kmer would be bases 1-47 of the read, the second kmer would be bases 2-48 of the read, the third kmer bases 3-49 and the fourth one bases 4-50, assuming you were also using a step size of 1.

Are you doing de novo assembly?

Ah ok.
It also said: Typically, K = 17.
So I guess they use 17 as a stanard length.

I am not doing it myself, we are doing it with the help of a company that does it for us.

But indeed, the goal is a de novo assembly.

[krobison]

17 is an arbitrary size. Depending on the assembler, the length of the reads, and their quality, quite frequently much higher values will give you better assembles.

Note that most assemblers prefer/demand odd length kmers, so that no kmer is its own reverse complement. Programs using kmers for other purposes may not impose that restriction.

[joskee]

17 is an arbitrary size. Depending on the assembler, the length of the reads, and their quality, quite frequently much higher values will give you better assembles.

Note that most assemblers prefer/demand odd length kmers, so that no kmer is its own reverse complement. Programs using kmers for other purposes may not impose that restriction.

Ok, I see what you mean.
I allready had that idea.
The odd numbered lenght kmers sounds logical! Good you told me, I would not have figured that out myself I think.

[biotech]

Note that most assemblers prefer/demand odd length kmers, so that no kmer is its own reverse complement. Programs using kmers for other purposes may not impose that restriction.

Could you explain this?
I do not see why it has to be an odd number.
Perhaps with a simple example.

[HESmith]

AAATTT: reverse complement is the same.
AAAATTT: reverse complement is AAATTTT.