def test_wr_sections_all():
"""Test that all sections files generated by wr_sections have the same content."""
f_sec_rd = "data/gjoneska_pfenning/sections_in.txt"
f_sec_wr = "tmp_test_sections_out.txt"
# Travis-CI path is cwd
f_sec_py = "tmp_test_sections.py"
# f_sec_mod = "tmp_test_sections"
# Read user GO IDs. Setup to write sections text file and Python file
usrgos = [getattr(nt, 'GO') for nt in goea_results]
sec_rd = _read_sections(f_sec_rd)
# Do preliminaries
godag = get_godag("go-basic.obo", prt=None, loading_bar=False, optional_attrs=['relationship'])
gosubdag = GoSubDag(usrgos, godag, relationships=True, tcntobj=None)
grprdflt = GrouperDflts(gosubdag)
# Exclude ungrouped "Misc." section of sections var(sec_rd)
hdrobj = HdrgosSections(gosubdag, grprdflt.hdrgos_dflt, sec_rd[:-1])
assert sec_rd[-1][0] == hdrobj.secdflt, sec_rd[-1][0]
grprobj = Grouper("test", usrgos, hdrobj, gosubdag)
# Create text and Python sections files
objsecwr = WrSectionsTxt(grprobj)
objsecwr.wr_txt_section_hdrgos(os.path.join(REPO, f_sec_wr))
objsecpy = WrSectionsPy(grprobj)
objsecpy.wr_py_sections(os.path.join(REPO, f_sec_py), sec_rd, doc=godag.version)
# Read text and Python sections files
sec_wr = _read_sections(f_sec_wr)
sec_py = _read_sections(f_sec_py)
def test_wr_sections_all():
"""Test that all sections files generated by wr_sections have the same content."""
f_sec_rd = os.path.join(REPO, "data/gjoneska_pfenning/sections_in.txt")
f_sec_wr = os.path.join(REPO, "tmp_test_sections_out.txt")
# Travis-CI path is cwd
f_sec_py = os.path.join(REPO, "tmp_test_sections.py")
# f_sec_mod = "tmp_test_sections"
# Read user GO IDs. Setup to write sections text file and Python file
usrgos = [getattr(nt, 'GO') for nt in goea_results]
sec_rd = _read_sections(f_sec_rd)
# Do preliminaries
godag = _get_godag()
gosubdag = GoSubDag(usrgos, godag, relationships=True, tcntobj=None)
grprdflt = _get_grprdflt(gosubdag)
# Exclude ungrouped "Misc." section of sections var(sec_rd)
hdrobj = HdrgosSections(gosubdag, grprdflt.hdrgos_dflt, sec_rd[:-1])
assert sec_rd[-1][0] == hdrobj.secdflt, sec_rd[-1][0]
grprobj = Grouper("test", usrgos, hdrobj, gosubdag)
# Create text and Python sections files
objsecwr = WrSectionsTxt(grprobj)
objsecwr.wr_txt_section_hdrgos(os.path.join(REPO, f_sec_wr))
objsecpy = WrSectionsPy(grprobj)
objsecpy.wr_py_sections(os.path.join(REPO, f_sec_py), sec_rd, doc=godag.version)
# Read text and Python sections files
sec_wr = _read_sections(f_sec_wr)
sec_py = _read_sections(f_sec_py)
def prt_txt(self, prt=sys.stdout):
"""Print an ASCII text format."""
prtfmt = self.objprtfmt.get_prtfmt_str(self.flds_cur)
prt.write("{FLDS}\n".format(FLDS=" ".join(self.flds_cur)))
WrSectionsTxt.prt_sections(prt,
self.desc2nts['sections'],
prtfmt,
secspc=True)
def prt_nts(self, desc2nts, prt=sys.stdout, prtfmt=None):
"""Print grouped and sorted GO IDs."""
# deprecated
# Set print format string
if prtfmt is None:
prtfmt = "{{hdr1usr01:2}} {FMT}\n".format(FMT=self.grprobj.gosubdag.prt_attr['fmt'])
# 1-D: data to print is a flat list of namedtuples
if 'flat' in desc2nts:
prt_txt(prt, desc2nts['flat'], prtfmt=prtfmt)
# 2-D: data to print is a list of [(section, nts), ...
else:
WrSectionsTxt.prt_sections(prt, desc2nts['sections'], prtfmt)
def prt_nts(self, desc2nts, prt=sys.stdout, prtfmt=None):
"""Print grouped and sorted GO IDs."""
# deprecated
# Set print format string
if prtfmt is None:
prtfmt = "{{hdr1usr01:2}} {FMT}\n".format(
FMT=self.grprobj.gosubdag.prt_attr['fmt'])
# 1-D: data to print is a flat list of namedtuples
if 'flat' in desc2nts:
prt_txt(prt, desc2nts['flat'], prtfmt=prtfmt)
# 2-D: data to print is a list of [(section, nts), ...
else:
WrSectionsTxt.prt_sections(prt, desc2nts['sections'], prtfmt)
def test_grouper_d2(do_plot=False):
"""Group depth-02 GO terms under their most specific depth-01 GO parent(s)."""
print('CWD', os.getcwd())
# Get GOs to be grouped
# Since no "Grouping GOs" were provided, depth-01 GOs are used for grouping.
grprdflt = _get_grprdflt()
hdrobj = HdrgosSections(grprdflt.gosubdag, grprdflt.hdrgos_dflt, sections=None, hdrgos=None)
grprobj = Grouper("Transient Increase", get_data0(), hdrobj, _get_gosubdag(), go2nt=None)
objwr = WrSectionsTxt(grprobj)
objwr.wr_txt_section_hdrgos("transient_increase_hdrgos.txt")
objwr.wr_txt_grouping_gos()
if do_plot:
# Don't run in Travis-CI because it does not contain 'dot'
from goatools.grouper.grprplt import GrouperPlot
GrouperPlot(grprobj).plot_groups_unplaced()
chk_hdrs(grprobj)
def _wr_sections_txt(fout_txt, usrgos, sections, grprdflt):
"""Given a list of usrgos and sections, write text file."""
try:
hdrobj = HdrgosSections(grprdflt.gosubdag, grprdflt.hdrgos_dflt, sections=sections)
grprobj = Grouper(fout_txt, usrgos, hdrobj, grprdflt.gosubdag, go2nt=None)
full_txt = os.path.join(REPO, fout_txt)
WrSectionsTxt(grprobj).wr_txt_section_hdrgos(full_txt, sortby=None, prt_section=True)
assert os.path.exists(full_txt)
except RuntimeError as inst:
sys.stdout.write("\n **FATAL: {MSG}\n\n".format(MSG=str(inst)))
def cli(self, prt=sys.stdout):
"""Command-line interface for go_draw script."""
kws = self.objdoc.get_docargs(prt=None)
godag = get_godag(kws['obo'], prt=None, loading_bar=False, optional_attrs=['relationship'])
usrgos = GetGOs(godag, max_gos=200).get_usrgos(kws.get('GO_FILE'), prt)
tcntobj = self._get_tcntobj(usrgos, godag, **kws) # Gets TermCounts or None
self.gosubdag = GoSubDag(usrgos, godag, relationships=True, tcntobj=tcntobj, prt=None)
grprdflt = GrouperDflts(self.gosubdag, kws['slims'])
ver_list = [godag.version, grprdflt.ver_goslims]
prt.write("{VER}\n".format(VER="\n".join(ver_list)))
sections = self._read_sections(kws['ifile'])
# print("SECSECSEC", sections)
hdrobj = HdrgosSections(self.gosubdag, grprdflt.hdrgos_dflt, sections)
grprobj = Grouper("init", usrgos, hdrobj, self.gosubdag)
# Write sections
objsecwr = WrSectionsTxt(grprobj, ver_list)
if not os.path.exists(kws['ifile']):
objsecwr.wr_txt_section_hdrgos(kws['ifile'])
objsecwr.wr_txt_section_hdrgos(kws['ofile'])
objsecpy = WrSectionsPy(grprobj, ver_list)
if 'py' in kws:
objsecpy.wr_py_sections(kws['py'], sections, doc=godag.version)
# Write user GO IDs in sections
sortobj = Sorter(grprobj)
objgowr = WrXlsxSortedGos("init", sortobj, ver_list)
objgowr.wr_txt_gos(kws['txt'], sortby=objsecpy.fncsortnt)
#objwr.wr_txt_section_hdrgos(kws['ofile'], sortby=objwr.fncsortnt)
self._prt_cnt_usrgos(usrgos, sys.stdout)
def test_grouper_d2(do_plot=False):
"""Group depth-02 GO terms under their most specific depth-01 GO parent(s)."""
# Get GOs to be grouped
# Since no "Grouping GOs" were provided, depth-01 GOs are used for grouping.
grprdflt = GrouperDflts()
hdrobj = HdrgosSections(grprdflt.gosubdag,
grprdflt.hdrgos_dflt,
sections=None,
hdrgos=None)
grprobj = Grouper("Transient Increase",
get_data0(),
hdrobj,
grprdflt.gosubdag,
go2nt=None)
objwr = WrSectionsTxt(grprobj)
objwr.wr_txt_section_hdrgos("transient_increase_hdrgos.txt")
objwr.wr_txt_grouping_gos()
if do_plot:
# Don't run in Travis-CI because it does not contain 'dot'
from goatools.grouper.grprplt import GrouperPlot
GrouperPlot(grprobj).plot_groups_unplaced()
chk_hdrs(grprobj)
def prt_txt(self, prt=sys.stdout):
"""Print an ASCII text format."""
prtfmt = self.objprtfmt.get_prtfmt_str(self.flds_cur)
prt.write("{FLDS}\n".format(FLDS=" ".join(self.flds_cur)))
WrSectionsTxt.prt_sections(prt, self.desc2nts['sections'], prtfmt, secspc=True)