But how do I convert my motif into meme format? I don't have any probabilities.
Hi I am trying to run fimo from the meme suite but can't make my sequence file readable.
I keep getting the error
> fimo motif ~/fastA.txt
There were no convincing matches to any MEME Suite motif format.
FATAL: No motifs could be read.
I am trying to run fimo with the motif file in text format. From the example file of meme I constructed my motif as a regular expression in such a form:
[ACTG][ACTG]TTC[CT][AG][CT]GAA[ACTG][ACTG]
Do I need to add more information to the file
Why can't fimo read my file?
thanks
Assa
1 answer
I think fimo doesn't explicitly take a regular expression. You have to create a meme file format like in this example:
MEME version 4
ALPHABET= ACGT
strands: + -
Background letter frequencies
A 0.303 C 0.183 G 0.209 T 0.306
MOTIF crp
letter-probability matrix: alength= 4 w= 19 nsites= 17 E= 4.1e-009
0.000000 0.176471 0.000000 0.823529
0.000000 0.058824 0.647059 0.294118
0.000000 0.058824 0.000000 0.941176
0.176471 0.000000 0.764706 0.058824
0.823529 0.058824 0.000000 0.117647
0.294118 0.176471 0.176471 0.352941
0.294118 0.352941 0.235294 0.117647
0.117647 0.235294 0.352941 0.294118
0.529412 0.000000 0.176471 0.294118
0.058824 0.235294 0.588235 0.117647
0.176471 0.235294 0.294118 0.294118
0.000000 0.058824 0.117647 0.823529
0.058824 0.882353 0.000000 0.058824
0.764706 0.000000 0.176471 0.058824
0.058824 0.882353 0.000000 0.058824
0.823529 0.058824 0.058824 0.058824
0.176471 0.411765 0.058824 0.352941
0.411765 0.000000 0.000000 0.588235
0.352941 0.058824 0.000000 0.588235
MOTIF lexA
letter-probability matrix: alength= 4 w= 18 nsites= 14 E= 3.2e-035
0.214286 0.000000 0.000000 0.785714
0.857143 0.000000 0.071429 0.071429
0.000000 1.000000 0.000000 0.000000
0.000000 0.000000 0.000000 1.000000
0.000000 0.000000 1.000000 0.000000
0.000000 0.000000 0.000000 1.000000
0.857143 0.000000 0.071429 0.071429
0.000000 0.071429 0.000000 0.928571
0.857143 0.000000 0.071429 0.071429
0.142857 0.000000 0.000000 0.857143
0.571429 0.071429 0.214286 0.142857
0.285714 0.285714 0.000000 0.428571
1.000000 0.000000 0.000000 0.000000
0.285714 0.214286 0.000000 0.500000
0.428571 0.500000 0.000000 0.071429
0.000000 1.000000 0.000000 0.000000
1.000000 0.000000 0.000000 0.000000
0.000000 0.000000 0.785714 0.214286
I guess you could assign 0 probability to the letters to exclude and equal probs to those to include. E.g. to represent [AG] you would put a row with "0.5 0 0.5 0" (where the alphabet is ACGT).
Yes this works. Thanks!
What do I do about the background frequencies? How are they being calculated here? Is it a separte background list I need to use or are they calculated from the frequencies of the list?
Log in to answer this question.