Genome arithmetics

The bnp.arithmetics module contains functions for analysing genomic intervals.

API documentation

bionumpy.arithmetics.forbes(chromosome_sizes: ChromosomeSize, intervals_a: Interval, intervals_b: Interval) → float[source]

Computes the Forbes similarity index for two sets of intervals.

Parameters

chromosome_sizesChromosomeSize: A ChromosomeSizes, typically from reading a chromosome.sizes file with bnp.open()
intervals_aInterval: Must be sorted. Can be read using bnp.open on a bed-file.
intervals_aInterval: Must be sorted. Can be read using bnp.open on a bed-file.

Returns

float: The forbes similarity index.

Examples

>>> from bionumpy.arithmetics import forbes, sort_intervals
>>> from bionumpy.datatypes import Interval
>>> a = Interval.from_entry_tuples([("chr1", 10, 20), ("chr2", 20, 30)])
>>> b = Interval.from_entry_tuples([("chr2", 15, 25), ("chr1", 10, 40)])
>>> a_sorted = sort_intervals(a, sort_order=["chr1", "chr2"])
>>> b_sorted = sort_intervals(b, sort_order=["chr1", "chr2"])
>>> forbes({"chr1": 100, "chr2": 200}, a_sorted, b_sorted)
5.625

bionumpy.arithmetics.jaccard(chromosome_sizes: ChromosomeSize, intervals_a: Interval, intervals_b: Interval) → float[source]

Computes the Jaccard similarity index for two sets of intervals.

Parameters

chromosome_sizesChromosomeSize: A ChromosomeSizes, typically from reading a chromosome.sizes file with bnp.open()
intervals_aInterval: Must be sorted. Can be read using bnp.open on a bed-file.
intervals_aInterval: Must be sorted. Can be read using bnp.open on a bed-file.

Returns

float: The forbes similarity index.

Examples

See forbes for examples.

bionumpy.arithmetics.sort_intervals(intervals: ~bionumpy.bnpdataclass.bnpdataclass.Interval, chromosome_key_function: callable = <function <lambda>>, sort_order: ~typing.List[str] | None = None) → Interval[source]

Sort intervals on “chromosome”, “start”, “stop”

Parameters

intervalsInterval: Unsorted intervals

Returns

Interval: Sorted intervals

bionumpy.arithmetics.merge_intervals(*args, **kwargs)

bionumpy.arithmetics.get_pileup(intervals: Interval, chromosome_size: int) → GenomicRunLengthArray[source]

Get the number of intervals that overlap each position of the chromosome/contig

This uses run length encoded arrays to handle the sparse data that we get from intervals.

Parameters

intervalsInterval,: Intervals on the same chromosome/contig
chromosome_sizeint: size of the chromsome/contig

Examples

>>> from bionumpy.datatypes import Interval
>>> from bionumpy.arithmetics import get_boolean_mask, get_pileup
>>> intervals = Interval(["chr1", "chr1", "chr1"], [3, 5, 10], [8, 7, 12])
>>> pileup = get_pileup(intervals, 20)
>>> print(pileup)
[0 0 0 1 1 2 2 1 0 0 1 1 0 0 0 0 0 0 0 0]

bionumpy.arithmetics.get_boolean_mask(intervals: Interval, chromosome_size: int)[source]

Get a boolean mask representing where any inteval hits

Uses run length encoded binary arrays to represent the areas covered by any interval. The mask that is returned supports numpy ufuncs, so that you can run logical operations on them s.a. & | ~ and also numpy indexing so you can use it to filter positions and intervals.

Parameters

intervalsInterval: Intervals on the same chromosome/contig
chromosome_sizeint: The size of the chromosome/contig

Examples

>>> intervals = Interval(["chr1", "chr1", "chr1"], [3, 5, 10], [8, 7, 12])
>>> print(intervals)
Interval with 3 entries
               chromosome                    start                     stop
                     chr1                        3                        8
                     chr1                        5                        7
                     chr1                       10                       12
>>> mask = get_boolean_mask(intervals, 20)
>>> print(mask.astype(int))
[0 0 0 1 1 1 1 1 0 0 1 1 0 0 0 0 0 0 0 0]

Get complement of the mask:

>>> complement = ~mask
>>> print(complement.astype(int))
[1 1 1 0 0 0 0 0 1 1 0 0 1 1 1 1 1 1 1 1]

Get the intersections (&) and union (|) of the mask and another mask

>>> other_mask = get_boolean_mask(Interval(["chr1"], [9], [15]), 20)
>>> intersection = mask & other_mask
>>> print(intersection.astype(int))
[0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0]
>>> union = mask | other_mask
>>> print(union.astype(int))
[0 0 0 1 1 1 1 1 0 1 1 1 1 1 1 0 0 0 0 0]

Find wether some positions overlap the mask: >>> print(other_mask[intervals.start]) [False False True]