def test_binnify():
chromsizes = bioframe.read_chromsizes(testdir +
"/test_data/test.chrom.sizes",
filter_chroms=False)
assert len(chromsizes) == 2
assert len(bioframe.binnify(chromsizes, int(np.max(
chromsizes.values)))) == len(chromsizes)
assert len(bioframe.binnify(chromsizes, int(np.min(
chromsizes.values)))) == (len(chromsizes) + 1)
assert len(bioframe.binnify(chromsizes, 1)) == np.sum(chromsizes.values)
def binnify(chromsizes_path, binsize, all_names):
import bioframe
chromsizes = bioframe.read_chromsizes(chromsizes_path,
filter_chroms=not (all_names))
bins = bioframe.binnify(chromsizes, binsize)
print(bins.to_csv(sep="\t", index=False))
def gene_content(genome, binsize, gc=True):
chrom_sizes = bioframe.fetch_chromsizes(genome)
chrom_table = binnify(chrom_sizes, binsize)
gene_count = frac_gene_coverage(chrom_table, genome)
if gc:
fasta_path = f'/net/levsha/share/lab/genomes/{genome}/{genome}.fa'
fasta_records = load_fasta(fasta_path)
gene_count['frac_gc'] = frac_gc(chrom_table, fasta_records)
return gene_count
def test_frac_mapped():
pytest.importorskip("pysam")
chromsizes = bioframe.read_chromsizes(testdir +
"/test_data/test.chrom.sizes",
filter_chroms=False)
fasta_records = bioframe.load_fasta(testdir + "/test_data/test.fa")
unmapped = np.array(
[1.0, 1.0, 1.0, 1.0, 0.0, 0.0, 1.0, 1.0, 1.0, 1.0, 0.0, 0.0])
assert (unmapped == bioframe.frac_mapped(bioframe.binnify(chromsizes, 1),
fasta_records,
return_input=False).values).all()
unmapped = np.array([0.8, 0.8, 0])
assert (unmapped == bioframe.frac_mapped(bioframe.binnify(chromsizes, 5),
fasta_records,
return_input=False).values).all()
unmapped = np.array([0.8, 4 / 7])
assert (unmapped == bioframe.frac_mapped(bioframe.binnify(chromsizes, 7),
fasta_records,
return_input=False).values).all()
def test_frac_gc():
pytest.importorskip("pysam")
chromsizes = bioframe.read_chromsizes(testdir +
"/test_data/test.chrom.sizes",
filter_chroms=False)
fasta_records = bioframe.load_fasta(testdir + "/test_data/test.fa")
unmapped_bp = (0 == bioframe.frac_mapped(bioframe.binnify(chromsizes, 1),
fasta_records,
return_input=False).values)
assert np.isnan(
bioframe.frac_gc(
bioframe.binnify(chromsizes, 1),
fasta_records,
return_input=False,
mapped_only=True,
).values[unmapped_bp]).all()
## mapped_only=True should ignore N or return np.nan if interval only contains N
np.testing.assert_equal(
np.array([0.5, 0.5, np.nan]),
bioframe.frac_gc(
bioframe.binnify(chromsizes, 5),
fasta_records,
return_input=False,
mapped_only=True,
).values,
)
assert (np.array([0.5, 0.5]) == bioframe.frac_gc(
bioframe.binnify(chromsizes, 7),
fasta_records,
return_input=False,
mapped_only=True,
).values).all()
## mapped_only=False should count N as zero
assert (np.array([0.4, 0.4, 0]) == bioframe.frac_gc(
bioframe.binnify(chromsizes, 5),
fasta_records,
return_input=False,
mapped_only=False,
).values).all()
assert (np.array([0.4, 2 / 7]) == bioframe.frac_gc(
bioframe.binnify(chromsizes, 7),
fasta_records,
return_input=False,
mapped_only=False,
).values).all()