ucsc-genomes-downloader 1.1.8

Simply instanziate a new genome

To download and load into memory the chromosomes of a given genomic assembly you can use the following code snippet:

from ucsc_genomes_downloader import Genome
hg19 = Genome(assembly="hg19")

Downloading selected chromosomes

If you want to select a subset of chromosomes to be downloaded you can use the attribute “chromosomes”:

from ucsc_genomes_downloader import Genome
hg19 = Genome("hg19", chromosomes=["chr1", "chr2"])

Getting gaps regions

The method return a DataFrame in bed-like format that contains the regions where only n or N nucleotides are present.

all_gaps = hg19.gaps() # Returns gaps (region formed of Ns) for all chromosomes
# Returns gaps for chromosome chrM
chrM_gaps = hg19.gaps(chromosomes=["chrM"])

Getting filled regions

The method return a DataFrame in bed-like format that contains the regions where no unknown nucleotide is present, basically the complementary of the gaps method.

all_filled = hg19.filled() # Returns filled for all chromosomes
# Returns filled for chromosome chrM
chrM_filled = hg19.filled(chromosomes=["chrM"])

Removing genome’s cache

To delete the cache of genome, including chromosomes and metadata you can use the delete method:

hg19.delete()

Genome objects representation

When printed, a Genome object has a human-readable representation. This allows you to print lists of Genome objects as follows:

print([
    hg19,
    hg38,
    mm10
])

# >>> [
#    Human, Homo sapiens, hg19, 2009-02-28, 25 chromosomes,
#    Human, Homo sapiens, hg38, 2013-12-29, 25 chromosomes,
#    Mouse, Mus musculus, mm10, 2011-12-29, 22 chromosomes
# ]

Obtaining a given bed file sequences

Given a pandas DataFrame in bed-like format, you can obtain the curresponding genomic sequences for the loaded assembly using the bed_to_sequence method:

my_bed = pd.read_csv("path/to/my/file.bed", sep="\t")
sequences = hg19.bed_to_sequence(my_bed)

Properties

A Genome object has the following properties:

hg19.assembly # Returns "hg19"
hg19.date # Returns "2009-02-28" as datetime object
hg19.organism # Returns "Human"
hg19.scientific_name # Returns "Homo sapiens"
hg19.description # Returns the brief description as provided from UCSC
hg19.path # Returns path where genome is cached

Retrieving a list of the available genomes

You can get a complete list of the genomes available from the UCSC website with the following method:

from ucsc_genomes_downloader.utils import get_available_genomes
all_genomes = get_available_genomes()

Tasselizing bed files

Create a tasselization of a given size of a given bed-like pandas dataframe.

Available alignment are to the left, right or center.

from ucsc_genomes_downloader.utils import tasselize_bed
import pandas as pd

my_bed = pd.read_csv("path/to/my/file.bed", sep="\t")
tasselized = tasselize_bed(
    my_bed,
    window_size=200,
    alignment="left"
)

Expand bed files regions

Expand a given dataframe in bed-like format using selected alignment.

Available alignment are to the left, right or center.

from ucsc_genomes_downloader.utils import expand_bed_regions
import pandas as pd

my_bed = pd.read_csv("path/to/my/file.bed", sep="\t")
expanded = expand_bed_regions(
    my_bed,
    window_size=1000,
    alignment="left"
)