region-selection

Methods for filtering for high-scoring genomic intervals

Usage

Importing the module and creating a Selection instance

>>> from region_selection import Selection
>>> s = Selection()

Specify properties

>>> s.method = "pq"
>>> s.input_fn = "/Users/areynolds/Developer/Github/region_selection/tests/windows.fixed.25k.bed"
>>> s.bin_size = 200
>>> s.exclusion_span = 24800

The method can be one of pq, wis, or maxmean, for selecting from one of priority-queue, weighted interval scheduling, or max-mean window sweep methods, respectively.

The input_fn property points to a file on the file system. This is optional, unless using the read() method.

The bin_size and exclusion_span properties are integers. These represent the size of elements, and the distance required between them (exclusing the bin, itself).

The default values are 200 and 24800, respectively. This means bins are 200 nt wide, and we require at least 25000 nt of distance between any filtered bins.

Input data

You can read in data from a four-column, tab-delimited text file:

>>> in_df = s.read(s, s.method, s.input_fn)
[region_selection] Reading input file into dataframe...
[region_selection] Read dataframe

Otherwise, you must provide a Pandas dataframe containing four columns, each labeled: Chromosome, Start, End, and Score, respecively.

In the above snippet, the input dataframe is called in_df.

Running the selection method

Use run() to run the specified method on the input dataframe in_df (or whatever its name is):

>>> out_df = s.run(s, s.method, in_df)
[region_selection] Bin size (nt): 200
[region_selection] Exclusion span (nt): 24800
[region_selection] Exclusion bins: 124
[region_selection] Method: Priority-Queue (PQ)
[region_selection] Constructing heap
[region_selection] Constructing qualifying bin list from heap
[region_selection] Returning sorted bin list
[region_selection] Method (runtime in sec): 140.50703937999998

The result is stored as a Pandas dataframe. Here it is called out_df and you can call all the usual Pandas properties on this:

>>> print(out_df.head())
    Chromosome   Start     End  Score
47        chr1    9400   34400   0.41
172       chr1   34400   59400   0.41
304       chr1   60800   85800   0.41
429       chr1   85800  110800   0.41
554       chr1  110800  135800   0.41

Or use the write() to write to standard output:

>>> s.write(out_df)
...

Or write with to_csv() etc.