create-priors [options] <FASTA file> <wiggle file>
The program create-priors takes as input a series of numeric values defined with respect to one or more DNA sequences. The program converts the data into a probabilistic prior using the method described in:
A binned distribution for the priors is also generated. Examples of input data types include sequence tags counts from a DNaseI hypersensitivity or histone modification ChIP-seq assay, or sequence conservation scores.
The prior and the prior distribution files can be used as supplementary input to FIMO and MCAST, improving the quality of their search results.A file in wiggle format containing the scores that will be converted to priors.
The sequences in the wiggle file should be a subset of the sequences included in the FASTA file. The wiggle file need not contain scores for every position in the FASTA file, but the sequences must be in the same order. Any positions missing from the input wiggle file will be assigned the median prior computed from the observed scores.create-priors will create a directory, named create-priors_out
by default. Any existing output files in the directory will be
overwritten. The directory will contain:
priors.wig. This file will contain the
position specific priors computed from the input tag counts.priors.dist. This file will
contain a probability distribution for the priors
(sample file).
The first three lines are the minimum, maximum, and median priors. Each
succeeding line contains the probability that a prior is in the
corresponding bin. The value of the prior is calculated from the bin
number using the formula
prior = ((bin #) / (num bins) * (max prior - min prior)) + (min prior)
The default output directory can be overridden using the --o or --oc options which are described below.
| Option | Parameter | Description | Default Behaviour |
|---|---|---|---|
| General Options | |||
| --alpha | num | The alpha parameter for calculating position specific priors. Alpha represents the fraction of all transcription factor binding sites that are binding sites for the TF of interest. Alpha must be between 0 and 1. | The default value is 1.0. |
| --beta | num | The beta parameter for calculating position specific priors. Beta represents an estimate of the total number of binding sites for all transcription factors in the input data. Beta must be greater than 0. | The default value is 10000. |
| --numbins | n | Number of bins to use in prior distribution file. | The default value is 100. |
| --parse-genomic-coord | When this option is specified each sequence header will be
checked for UCSC style genomic coordinates. These are of the form:
>sequence name:starting position-ending position
Where
|
The first position in the sequence will be assumed to be 1. | |
| --psp | Directs create-priors to also output the priors in
MEME PSP format.
The name of the output file will be create-priors.psp.
This format is only suitable when the number of positions is
relatively small. |
||