How Can I Re-Format My Dna Motifs' Position Weight Matrices (Pwms)?

Question

[a1ultima]

I am working with a set of DNA motifs that are predicted as potential regulatory motifs (e.g. transcription factor binding sites). The motifs belong to several species, and I wanted to cluster these motifs via their Position Weight Matrices (PWMs) (also known as PSSMs) to collapse similar motifs together into groups.

A tool called MATLIGN (website here) does what I need, but their required format for the PWMs are different to what I have, they claim:

"Matrices must be in the frequency matrix format (only integer numbers are acceptable)"

The problem is that my PWM matrices do not have integer numbers but decimals instead. e.g.:

     A        C        G        T
1    0.000000 1.000000 0.000000 0.000000
2    1.000000 0.000000 0.000000 0.000000
3    0.000000 0.000000 1.000000 0.000000
4    0.000000 0.421755 0.000000 0.578245
5    0.289407 0.000000 0.282556 0.428038

In other words, instead of the decimal values I have in my matrix I need to have integer counts. Could anybody suggest what I can do? Would I need to create artificial counts?

[UnivStudent]

That looks a lot like a position frequency matrix (PFM) where the counts were divided by the row total. Unless you know that this had a background nucleotide frequency taken into account you can probably just multiply everything by a constant and round to make it into 'counts'. You can also use a tool like TOMTOM to do this where it doesn't require integers.

[a1ultima]

@UnivStudent: I have actually used TOMTOM before, unfortunately they only do pairwise motif comparisons. I was hoping to use a more advanced method that carries out clustering as well.

[Asaf]

PWM contain less information than the actual counts. Where did you obtain the PWM from? Try to find the counts or actual sequences as well.

Answer 1

[a1ultima]

After looking more closely at my data, I noticed that with each of the PWMs I had there was a value called nSites.

It turns out that nSites refers to the number of DNA sequence regions, or sites, that were used to originally generate the PWMs.

Solution:

With this I was able to convert my PWMs into integer counts by multiplying the proportions by the nSites value.