Setting/finding the log odds detection threshold in custom homer PWMs

Question

[endrebak]

I want to convert the JASPAR pwms to homer format:

http://jaspar.genereg.net/downloads/

It seems like the JASPAR files in the transfac format are pretty close:

AC MA0004.1
XX
ID Arnt
XX
DE MA0004.1 Arnt ; From JASPAR 2018
PO  A   C   G   T
01  4.0 16.0    0.0 0.0
02  19.0    0.0 1.0 0.0
03  0.0 20.0    0.0 0.0
04  0.0 0.0 20.0    0.0
05  0.0 0.0 0.0 20.0
06  0.0 0.0 20.0    0.0
XX

To compare, this is what the homer format looks like:

>ATGACTCATC     AP-1(bZIP)/ThioMac-PU.1-ChIP-Seq(GSE21512)/Homer        6.049537        -1.782996e+03   0       9805.3,5781.0,3085.1,2715.0,0.00e+00
0.419   0.275   0.277   0.028
0.001   0.001   0.001   0.997
0.010   0.002   0.965   0.023
0.984   0.003   0.001   0.012
0.062   0.579   0.305   0.054
0.026   0.001   0.001   0.972
0.043   0.943   0.001   0.012
0.980   0.005   0.001   0.014
0.050   0.172   0.307   0.471
0.149   0.444   0.211   0.195

However, I need to find the value for the Log odds detection threshold (third column in header) myself. How do I do that?

More info on creating custom matrixes in homer here: http://homer.ucsd.edu/homer/motif/creatingCustomMotifs.html

[tiago211287]

good unanswered question

Answer 1

[jared.andrews07]

I created a python wrapper around a program called TFMpvalue a few years ago that does just that. I am not continuing support/development for it, but it may be useful for this. In short, it will generate a score threshold for a given p-value for a PWM and vice-versa.