def one_to_one_tuples(*,
kits: List[str],
output_prefix: Path
) -> List[Optional[OneToOneAutosomeResult]]:
'''Performs one-to-one autosomal analysis.
Args:
kits: one or more kit name or number tuples (comma seperated).
output_prefix: the path prefix for the output files.
'''
kits_dict: Dict[str, Kit] = dict([(kit.number, kit) for kit in find_kits()])
results: List[Optional[OneToOneAutosomeResult]] = []
logging.info(f'processing {len(kits)} kit pairs.')
for i, kit_tuple in enumerate(kits, 1):
kit_one, kit_two = kit_tuple.split(',')
logging.info(f'processing ({i}/{len(kits)}) kits {kit_one} and {kit_two}.')
tuple_output_prefix = Path(str(output_prefix) + f'.{kit_one}-{kit_two}')
result = one_to_one_api(kit_one=kit_one,
kit_two=kit_two,
output_prefix=tuple_output_prefix,
kits=kits_dict)
if result is None:
logging.warning(f'No 1:1 autosomal match found for kits {kit_one} and {kit_two}.')
results.append(result)
logging.info(f'writing output.')
summary: Path = Path(str(output_prefix) + '.summary.txt')
write_metrics(summary, results)
return results
def _one_to_many(
kit: str,
output: Path,
max_matches: Optional[int],
kits: Optional[Dict[str, Kit]] = None,
driver: Optional[WebDriver] = None) -> List['OneToManyAutosomeResult']:
'''Performs one-to-many autosomal analysis.
Args:
kit: the name or number
output: the output file.
max_matches: the maximum # of matches to return
kits: a mapping of kit name to kit, useful when performing many 1:1 analyses.
Returns:
A list of matches.
'''
results: List['OneToManyAutosomeResult'] = []
_driver = main_page() if driver is None else driver
if kits is None:
kits = dict([(kit.number, kit) for kit in _ls(_driver)])
if kit not in kits:
kit = [n for n, k in kits.items() if k.name == kit][0]
try:
url = f'OneToMany0Tier2.php?kit_num={kit}'
page = _driver.find_element_by_xpath('//a[@href="' + url + '"]')
page.click()
root = lxml.html.fromstring(_driver.page_source)
for table in root.xpath("//table"):
rows = [row for row in table.xpath('.//tr')]
# check that the first column in the first row has value "Chr"
tds = rows[0].xpath('.//td')
first_column_value = tds[0].text
if first_column_value != 'Kit':
continue
header = [td.text for td in tds]
assert len(header) == 10, f'header: {header}'
last_row: int = len(
rows) if max_matches is None else max_matches + 1
logging.info(f'Reading rows for {kit}')
for row_num, row in enumerate(rows[1:last_row], 1):
columns = [maybe_href(td) for td in row.xpath('.//td')]
assert len(columns) == 10, f'columns: {columns}'
d = dict(zip(header, columns))
result = OneToManyAutosomeResult(
kit_one=kits[kit],
kit_two=Kit(name=d['Name'],
number=d['Kit'],
email=d['Email'],
testing_company=d['Testing Company']),
largest_segment=float(d['Largest Seg']),
total_half_match_segments=float(d['Total cM']),
most_recent_common_ancestor=float(d['Gen']),
num_snps=int(d['Overlap']),
date_compared=d['Date Compared'])
results.append(result)
logging.info(f'Returning {len(results)} results.')
except Exception as e:
_driver.close()
raise e
if driver is None:
_driver.close()
write_metrics(output, results)
return results
def _one_to_one(
kit_one: str,
kit_two: str,
output_prefix: Path,
kits: Optional[Dict[str, Kit]] = None,
driver: Optional[WebDriver] = None
) -> Optional['OneToOneAutosomeResult']:
'''Performs one-to-one autosomal analysis.
Args:
kit_one: the first kit name or number
kit_two: the second kit name or number
output_prefix: the prefix for the output files.
kits: a mapping of kit name to kit, useful when performing many 1:1 analyses.
Returns:
None if the analysis did not find any segments, otherwise the analysis results.
'''
_driver = main_page() if driver is None else driver
if kits is None:
kits = dict([(kit.number, kit) for kit in _ls(_driver)])
if kit_one not in kits:
kit_one = [n for n, k in kits.items() if k.name == kit_one][0]
if kit_two not in kits:
kit_two = [n for n, k in kits.items() if k.name == kit_two][0]
try:
url = 'v_compare1.php'
page = _driver.find_element_by_xpath('//a[@href="' + url + '"]')
page.click()
kit1 = _driver.find_element_by_name('kit1')
kit1.clear()
kit1.send_keys(kit_one)
kit2 = _driver.find_element_by_name('kit2')
kit2.clear()
kit2.send_keys(kit_two)
submit = _driver.find_element_by_name('xsubmit')
submit.click()
segments: List[SegmentResult] = []
root = lxml.html.fromstring(_driver.page_source)
for table in root.xpath("//table"):
rows = [row for row in table.xpath('.//tr')]
# check that the first column in the first row has value "Chr"
tds = rows[0].xpath('.//td')
first_column_value = tds[0].text
if first_column_value != 'Chr':
continue
for row in rows[1:]:
columns = row.xpath('.//td')
assert len(columns) == 5
segment = SegmentResult(
chromosome=columns[0].text,
start=_comma_value_to_int(columns[1].text),
end=_comma_value_to_int(columns[2].text),
centimorgans=float(columns[3].text),
num_snps=_comma_value_to_int(columns[4].text))
segments.append(segment)
largest_segment: float = 0.0
total_half_match_segments: float = 0.0
pct_half_match_segments: float = 0.0
most_recent_common_ancestor: float = -1.0
shared_segments: int = 0
num_snps: int = 0
pct_snps_identical: float = 0.0
version: str = 'none found'
for line in _driver.page_source.split('\n'):
line = line.rstrip('\r\n').strip()
line = line.replace('<br>', '')
if 'No shared DNA segments found' in line:
assert len(segments) == 0
elif 'Largest segment' in line:
largest_segment = float(line.split(' = ')[1].split(' ')[0])
elif 'Total Half-Match segments' in line:
fields = line.split(' = ')[1].split(' ')
total_half_match_segments = float(fields[0])
pct_half_match_segments = float(fields[2].replace('(',
'').replace(
')', ''))
elif 'Estimated number of generations to MRCA' in line:
most_recent_common_ancestor = float(
line.split(' = ')[1].split(' ')[0])
elif 'shared segments found for this comparison' in line:
shared_segments = int(line.split(' ')[0])
elif 'SNPs used for this comparison' in line:
num_snps = int(line.split(' ')[0])
elif 'Pct SNPs are full identical' in line:
pct_snps_identical = float(line.split(' ')[0])
elif 'Ver:' in line:
version = line.replace('<font size="2">Ver: ', '') \
.replace('</font>', '') \
.replace(' ', '-')
final_result = None
vars = [
largest_segment, total_half_match_segments,
pct_half_match_segments, most_recent_common_ancestor,
shared_segments, num_snps, pct_snps_identical, version
]
if all([v is not None for v in vars]):
final_result = OneToOneAutosomeResult(
kit_one=kits[kit_one],
kit_two=kits[kit_two],
segments=segments,
largest_segment=largest_segment,
total_half_match_segments=total_half_match_segments,
pct_half_match_segments=pct_half_match_segments,
most_recent_common_ancestor=most_recent_common_ancestor,
shared_segments=shared_segments,
num_snps=num_snps,
pct_snps_identical=pct_snps_identical,
version=version)
except Exception as e:
_driver.close()
raise e
if driver is None:
_driver.close()
if final_result is not None:
summary: Path = Path(str(output_prefix) + '.summary.txt')
write_metric(summary, final_result)
detailed: Path = Path(str(output_prefix) + '.detailed.txt')
write_metrics(detailed, final_result.segments)
return final_result