def test_annotation_init(caplog, annot):
assert annot.genome_file.endswith("data/regexp/regexp.fa")
assert annot.annotation_gtf_file.endswith("data/regexp/regexp.annotation.gtf")
assert annot.annotation_bed_file.endswith("data/regexp/regexp.annotation.bed")
assert annot.annotation_contigs == ["chrM"] # from BED
assert len(annot.genome_contigs) == 17 # from sizes
assert isinstance(annot.bed, pd.DataFrame)
assert isinstance(annot.gtf, pd.DataFrame)
assert annot.genes("bed") == ["NP_059343.1"]
# >1 GTF files: take unzipped
g1 = "tests/data/data.annotation.gtf"
with genomepy.files._open(g1, "w") as fa:
fa.write("1")
g2 = "tests/data/data.annotation.gtf.gz"
with genomepy.files._open(g2, "w") as fa:
fa.write("2")
a = genomepy.Annotation("data", genomes_dir="tests")
assert a.annotation_gtf_file.endswith(g1)
genomepy.utils.rm_rf(g1)
genomepy.utils.rm_rf(g2)
# not enough GTF files
genomepy.Annotation("empty", genomes_dir="tests/data")
assert "Could not find 'empty.annotation.bed" in caplog.text
assert "Could not find 'empty.annotation.gtf" in caplog.text
# Genome doesn't exist
with pytest.raises(FileNotFoundError):
genomepy.Annotation("never_existed", genomes_dir="tests/data")
def test_named_gtf():
a = genomepy.Annotation("sacCer3", genomes_dir="tests/data")
df = a.named_gtf
assert df.index.name == "gene_name"
assert str(a.gtf.index.dtype) == "int64"
assert str(df.index.dtype) == "object"
assert set(df.at["YDL248W", "seqname"]) == {"chrIV"}
def test_match_contigs():
# nothing to work with
a = genomepy.Annotation("sacCer3", genomes_dir="tests/data")
cd = genomepy.annotation.sanitize._match_contigs(a)
assert cd is None
# one missing contig, one fixable contig
a = genomepy.Annotation("sanitize", genomes_dir="tests/data")
before = a.gtf
cd = genomepy.annotation.sanitize._match_contigs(a)
after = a.gtf
assert cd == {"NC_007112.7": "1"}
assert a.genome_contigs == ["1"]
assert list(before.seqname.unique()) == ["NC_007112.7", "NC_002333.2"]
assert list(after.seqname.unique()) == ["1", "NC_002333.2"]
def test_custom_annotation():
for fname in [
"tests/data/custom.annotation.bed",
"tests/data/custom.annotation.bed.gz",
]:
a = genomepy.Annotation(name=fname, genomes_dir="tests/data")
assert a.bed.shape[0] == 10
for fname in [
"tests/data/custom.annotation.gtf",
"tests/data/custom.annotation.gtf.gz",
]:
a = genomepy.Annotation(name=fname, genomes_dir="tests/data")
assert a.gtf.shape[0] == 45
with pytest.raises(NotImplementedError):
a = genomepy.Annotation(name="tests/data/regions.txt", genomes_dir="tests/data")
def test_filter_contigs():
a = genomepy.Annotation("sacCer3", genomes_dir="tests/data")
assert "chrV" not in a.bed.chrom.unique()
# add a chromosome to the BED not present in the genome
a.bed.at[0, "chrom"] = "chrV"
assert "chrV" in a.bed.chrom.unique()
missing_contigs = genomepy.annotation.sanitize._filter_contigs(a)
assert missing_contigs == {"chrV"}
assert "chrV" not in a.bed.chrom.unique()
def test__parse_annot():
a = genomepy.Annotation("sacCer3", genomes_dir="tests/data")
df = genomepy.annotation.utils._parse_annot(a, "bed")
assert df.equals(a.bed)
df = genomepy.annotation.utils._parse_annot(a, "gtf")
assert df.equals(a.gtf)
df = genomepy.annotation.utils._parse_annot(a, a.bed)
assert df.equals(a.bed)
with pytest.raises(ValueError):
genomepy.annotation.utils._parse_annot(a, 1)
def test_map_locations():
a = genomepy.Annotation("sacCer3", genomes_dir="tests/data")
# BED + Ensembl
ens = a.map_locations(annot=a.bed.head(1), to="Ensembl")
assert ens.chrom.to_list() == ["IV"]
# GTF + NCBI
ncbi = a.map_locations(annot=a.gtf.head(1), to="NCBI")
assert ncbi.seqname.to_list() == ["IV"]
# custom dataframe (already indexed)
df = a.bed.head(1).set_index("start")
cus = a.map_locations(annot=df, to="Ensembl")
assert ens.chrom.to_list() == ["IV"]
assert cus.index.name == "start"
def test_gene_coords(caplog):
a = genomepy.Annotation("sacCer3", genomes_dir="tests/data")
bed_genes = a.genes("bed")[0:10]
c = a.gene_coords(bed_genes, "bed")
assert list(c.shape) == [10, 5]
assert c.columns.to_list() == ["chrom", "start", "end", "name", "strand"]
gtf_genes = a.genes("gtf")[0:10]
c = a.gene_coords(gtf_genes, "gtf")
assert list(c.shape) == [10, 5]
assert c.columns.to_list() == ["seqname", "start", "end", "gene_name", "strand"]
# mismatched gene names!
_ = a.gene_coords(["what?"], "bed")
assert "No genes found." in caplog.text
_ = a.gene_coords(["YDL200C_mRNA", "what?"], "bed")
assert "Only 50% of genes was found." in caplog.text
def test_map_genes():
a = genomepy.Annotation("GRCz11", genomes_dir="tests/data")
bed = a.bed.head()
transcript_ids = bed.name.to_list()
assert transcript_ids[0] == "ENSDART00000159919"
# transcript to gene
res = a.map_genes(field="ensembl.gene", annot=bed)
genes = res.name.to_list()
assert genes[0] == "ENSDARG00000103202"
# # transcript to symbol
# res = a.map_genes(field="symbol", df=bed)
# symbol = res.name.to_list()
# assert symbol[0] == "CR383668.1"
# refseq hits & subtypes
protein = a.map_genes(field="refseq", product="protein", annot=bed)
assert protein.name.to_list()[0].startswith("NP_")
def test_sanitize():
a = genomepy.Annotation("sanitize", genomes_dir="tests/data")
a.sanitize(overwrite=False)
assert a.gtf.shape == (4, 9)
assert list(set(a.gtf.seqname)) == ["1"]
def test_genes():
a = genomepy.Annotation("GRCz11", genomes_dir="tests/data")
g = a.genes("bed")
assert isinstance(g, list)
g = a.genes("gtf")
assert isinstance(g, list)
ax.set_xlabel("contig number (ordered by size)")
ax.set_ylabel("contig size")
ax.set_ylim((1, 15 * max(sizes)))
plt.xscale("log")
plt.yscale("log")
# now save the image as html text
img = io.BytesIO()
fig.savefig(img, format='png')
img.seek(0)
html = '<img src="data:image/png;base64, {}">'.format(
base64.b64encode(img.getvalue()).decode('utf-8'))
# if we have an annotation check for the number of genes present
if hasattr(snakemake.input, "annotation"):
gp = genomepy.Annotation(assembly,
genomes_dir=snakemake.params.genomes_dir)
nr_genes = len(gp.genes("gtf"))
annotation_text = f"""The genome annotation contains {nr_genes} genes."""
else:
annotation_text = ""
# save it all
with open(outfile, "a") as f:
f.write(f"""
<!--
id: 'assembly_stats'
section_name: 'Assembly stats'
-->
Genome assembly {assembly} contains of {len(sizes)} contigs, with a GC-content \
of {gc / (gc + at) * 100:.2f}%, and {(total_size - gc - at) / total_size * 100:.2f}%\
consists of the letter N. The <a href="https://en.wikipedia.org/wiki/N50,_L50,_and_r\
