Hi Bruno,

The significance cutoff you choose all depends on what you are trying
to do.  A more stringent cutoff will likely have a higher proportion
of true functional sites, but this will come at the cost of missing
true sites that might not be as strongly conserved.  Likewise, a less
stringent cutoff will capture more true binding sites, but will also
introduce more false positives.  Regardless, it is important to keep
in mind that these are simply predictions based on conserved
sequences.

As for converting from Z-scores to P-values, this can be achieved
using the normal cumulative probability distribution.  This is
implemented in most statistical packages like R, or you can use a
lookup table like one:
http://www.epatric.com/documentation/statistics/z-score_table.html

Matt
(re-post)

Bruno Pardo wrote:
> Hello,
>
> I am using  the UCSC genome Browser to find
> informations about transcription factors binding sites.
>
> I retrieved the Data from  the tfbsConsSites Table
> (group regulation, track TFBS Conserved).
>
> I would like to know if the sites with a z-score <2.33
> can be considered as sufficiently significant and
> how to calculate the pvalue from the z-score.
>
> Best Regards
>
> Bruno Pardo
> _______________________________________________
> Genome maillist  -  [email protected]
> https://lists.soe.ucsc.edu/mailman/listinfo/genome
>   
_______________________________________________
Genome maillist  -  [email protected]
https://lists.soe.ucsc.edu/mailman/listinfo/genome

Reply via email to