Full sequence output from AME

[XIYIN SHI]

Hi,

I used AME for known motif enrichment analysis on a set of sequences, and I noticed that the `sequences.tsv` output only includes sequences identified as positives — those scoring above a threshold.

Since I'm interested in obtaining PWM scores for all input sequences, including those below the threshold, I'm wondering is there a way to configure AME to output scores for all sequences, or to change/disable the default threshold?

Thanks!

Xiyin

[cegrant]

I'm not quite sure what you are looking for here, or if AME is the appropriate tool. The primary function of AME is to look for motif enrichment in a collection of sequences that are scored by some metric.  In the absence of sequence metrics AME will simply rank the sequences by their position in the file. If you don't have some biological metric ranking the sequences, and are simply looking for evidence of motif enrichment, you should consider using SEA instead. You should study the AME command line documentation to make sure it's the appropriate tool for your analysis. For your particular request, look at the --hit-lo-fraction  option. You can't set this to 0, but if you set it to a sufficiently small number, all of your sequences should pass the threshold.

If you are primarily interested in PWM scores you might consider using FIMO rather than AME. FIMO's only function is to identify matches to the PWM that pass a statistical confidence threshold.