snplib 1.0.0__py3-none-any.whl

format/_plink.py

@@ -0,0 +1,305 @@
+#!/usr/bin/env python
+# coding: utf-8
+__author__ = "Igor Loschinin (igor.loschinin@gmail.com)"
+
+__all__ = [
+    "make_map", "make_ped", "make_fam", "make_lgen"
+]
+
+import re
+import pandas as pd
+
+
+def make_map(manifest: pd.DataFrame) -> pd.DataFrame:
+    """ PLINK text fileset variant information file
+    https://www.cog-genomics.org/plink/1.9/formats#map
+
+    A text file with no header line, and one line per variant with the following 3-4 fields:
+
+        1. Chromosome code. PLINK 1.9 also permits contig names here, but most
+            older programs do not.
+        2. Variant identifier
+        3. Position in morgans or centimorgans (optional; also safe to use
+            dummy value of '0')
+        4. Base-pair coordinate
+
+        All lines must have the same number of columns (so either no lines
+    contain the morgans/centimorgans column, or all of them do).
+
+    :param manifest: The file that is taken on the Illumina website with full
+    information about the chip
+    https://support.illumina.com/downloads/bovinesnp50-v3-0-product-files.html
+
+    :return: Return data in formate .map
+    """
+
+    fields = ['Chr', 'Name', 'MapInfo']
+
+    if all([
+        True
+        if item not in manifest.columns
+        else False
+        for item in fields
+    ]):
+        raise KeyError("Manifest has no data to build map format!")
+
+    # Rearrange the columns and replace the names of the sex and mitochondrial
+    # chromosomes
+    permute_cols = manifest[fields].\
+        sort_values(by='Name').\
+        replace({'X': 30, 'Y': 31, 'MT': 33}).\
+        dropna(axis=0)
+
+    # Insert distances in centimorganides
+    permute_cols.insert(2, 'morgans', [0] * len(manifest))
+
+    return permute_cols
+
+
+def make_ped(
+        data: pd.DataFrame,
+        sid_col: str,
+        snp_col: str,
+        fid_col: str = None,
+        father_col: str = None,
+        mother_col: str = None,
+        sex_col: str = None,
+) -> pd.DataFrame | None:
+    """ Original standard text format for sample pedigree information and
+    genotype calls. Normally must be accompanied by a .map file.
+    https://www.cog-genomics.org/plink/1.9/formats#ped
+
+        The PED file has 6 fixed columns at the beginning followed by the SNP
+    information. The columns should be separated by a whitespace or a tab. The
+    first six columns hold the following information:
+
+        1. Family ID (if unknown use the same id as for the sample id in
+            column two)
+        2. Sample ID
+        3. Paternal ID (if unknown use 0)
+        4. Maternal ID (if unknown use 0)
+        5. Sex (1=male; 2=female; 0=unknown)
+        6. Affection (0=unknown; 1=unaffected; 2=affected)
+        7. Genotypes (space or tab separated, 2 for each marker. 0/-9=missing)
+
+        Here is a brief example of a genotype PED file containing 5 samples
+    with 10 homozygous SNPs:
+        4304 4304 0 0 0 0 C C C C G G G G G G C C G G C C T T T T
+        6925 6925 0 0 0 0 C C C C T T G G A A C C G G C C T T T T
+        7319 7319 0 0 0 0 C C C C G G G G G G C C G G C C T T T T
+        6963 6963 0 0 0 0 A A C C T T G G A A C C G G C C T T T T
+        6968 6968 0 0 0 0 C C C C G G G G G G G G G G C C T T T T
+
+    :param data: Snp data that contain full or partial information on the
+        animal
+    :param sid_col: Sample ID. Column name in data
+    :param snp_col: Snp column name in data
+    :param fid_col: Family ID column name in data (if unknown use the same
+        id as for the sample id in column two)
+    :param father_col: Paternal ID column name in data (if unknown use 0)
+    :param mother_col: Maternal ID column name in data (if unknown use 0)
+    :param sex_col: Sex column name in data (if unknown use 0)
+    :return: Returns an array of data in ped format to work with the plink
+        program
+    """
+
+    _fields = ["fid", "sid", "father", "mother", "sex", "not_used", "snp"]
+    _f_dtype = dict(zip(_fields, (str for _ in range(len(_fields)))))
+
+    _ped = pd.DataFrame(columns=_fields)
+
+    if sid_col not in data.columns or snp_col not in data.columns:
+        raise KeyError(f"Data has not in name columns!")
+
+    # Checked Sample ID on underscope - '_'
+    _ped["sid"] = data[sid_col].astype(str)
+    if _ped["sid"].apply(_check_underscore).any():
+        raise Exception(
+            "Replace in 'Sample ID' columns '_' on another a simbols"
+        )
+
+    # Checked Family ID on underscope - '_'
+    if fid_col is not None:
+        if fid_col not in data.columns:
+            raise KeyError(f"Data has not in name columns {fid_col}!")
+
+        if (data[fid_col].dtype.hasobject and
+                data[fid_col].apply(_check_underscore).any()):
+            raise Exception(
+                "Replace in 'Family ID' columns '_' on another a simbols"
+            )
+
+        _ped["fid"] = data[fid_col]
+
+    else:
+        _ped["fid"] = data[sid_col].astype(str)
+
+    _ped["father"] = data[father_col] if father_col is not None else 0
+    _ped["mother"] = data[mother_col] if mother_col is not None else 0
+    _ped["sex"] = data[sex_col] if sex_col is not None else 0
+    _ped["not_used"] = 0
+    _ped["snp"] = data[snp_col]
+
+    return _ped[_fields].astype(_f_dtype)
+
+
+def make_fam(
+        data: pd.DataFrame,
+        sid_col: str,
+        fid_col: str = None,
+        father_col: str = None,
+        mother_col: str = None,
+        sex_col: str = None,
+        sex_val: int = 0,
+        pheno_col: str = None,
+        pheno_val: int = -9
+
+) -> pd.DataFrame | None:
+    """ PLINK sample information file
+    https://www.cog-genomics.org/plink/1.9/formats#fam
+
+        A text file with no header line, and one line per sample with the
+    following six fields:
+
+        1. Family ID ('FID')
+        2. Within-family ID ('IID'; cannot be '0')
+        3. Within-family ID of father ('0' if father isn't in dataset)
+        4. Within-family ID of mother ('0' if mother isn't in dataset)
+        5. Sex code ('1' = male, '2' = female, '0' = unknown)
+        6. Phenotype value ('1' = control, '2' = case, '-9'/'0'/non-numeric =
+            missing data if case/control)
+
+    :param data: Snp data that contain full or partial information on the
+        animal
+    :param fid_col: Family ID, default value "1". Must not contain
+        underline - "_"
+    :param sid_col: Within-family ID ('IID'; cannot be '0'). Must not contain
+        underline - "_"
+    :param father_col: Within-family ID of father ('0' if father isn't in
+        dataset)
+    :param mother_col: Within-family ID of mother ('0' if mother isn't in
+        dataset)
+    :param sex_col: Sex column name in data
+    :param sex_val: Sex code ('1' = male, '2' = female, '0' = unknown)
+    :param pheno_col: Pheno column name in data
+    :param pheno_val: Phenotype value ('1' = control, '2' = case,
+        '-9'/'0'/non-numeric = missing data if case/control)
+    :return: Return data in formate .fam
+    """
+
+    _fields = ['fid', 'sid', 'father', 'mother', 'sex', 'pheno']
+    _f_dtype = dict(zip(_fields, (str for _ in range(len(_fields)))))
+
+    _fam = pd.DataFrame(columns=_fields)
+
+    if sid_col not in data.columns:
+        raise KeyError(f"Data has not in name columns {sid_col}!")
+
+    # Checked Sample ID on underscope - '_'
+    _fam["sid"] = data[sid_col].astype(str)
+    if _fam["sid"].apply(_check_underscore).any():
+        raise Exception(
+            "Replace in 'Sample ID' columns '_' on another a simbols"
+        )
+
+    # Checked Family ID on underscope - '_'
+    if fid_col is not None:
+        if fid_col not in data.columns:
+            raise KeyError(f"Data has not in name columns {fid_col}!")
+
+        if (data[fid_col].dtype.hasobject and
+                data[fid_col].apply(_check_underscore).any()):
+            raise Exception(
+                "Replace in 'Family ID' columns '_' on another a simbols"
+            )
+
+        _fam["fid"] = data[fid_col]
+
+    else:
+        _fam["fid"] = 1
+
+    _fam["father"] = data[father_col] if father_col is not None else 0
+    _fam["mother"] = data[mother_col] if mother_col is not None else 0
+    _fam["sex"] = data[sex_col] if sex_col is not None else sex_val
+    _fam['pheno'] = data[pheno_col] if pheno_col is not None else pheno_val
+
+    return _fam[_fields].astype(_f_dtype)
+
+
+def make_lgen(
+        data: pd.DataFrame,
+        sid_col: str,
+        snp_name: str,
+        alleles: list[str],
+        fid_col: str = None
+) -> pd.DataFrame | None:
+    """ PLINK long-format genotype file
+    https://www.cog-genomics.org/plink/1.9/formats#lgen
+
+        A text file with no header line, and one line per genotype call (or
+    just not-homozygous-major calls if 'lgen-ref' was invoked) usually with
+    the following five fields:
+
+        1. Family ID
+        2. Within-family ID
+        3. Variant identifier
+        4. Allele call 1 ('0' for missing)
+        5. Allele call 2
+
+        There are several variations which are also handled by PLINK; see the
+    original discussion for details.
+
+    :param data: Data the after parsing FinalReport.txt
+    :param sid_col:
+    :param snp_name:
+    :param fid_col: Family ID, default value "1"
+    :param alleles:
+    :return: - Return data in formate .lgen
+    """
+    _fields = ['fid', 'sid', 'snp_name', 'allele1', 'allele2']
+    _f_dtype = dict(zip(_fields, (str for _ in range(len(_fields)))))
+
+    _lgen = pd.DataFrame(columns=_fields)
+
+    try:
+        # Checked Sample ID on underscope - '_'
+        _lgen["sid"] = data[sid_col].astype(str)
+        if _lgen["sid"].apply(_check_underscore).any():
+            raise Exception(
+                "Replace in 'Sample ID' columns '_' on another a simbols"
+            )
+
+        # Checked Family ID on underscope - '_'
+        if fid_col is not None:
+            if (data[fid_col].dtype.hasobject and
+                    data[fid_col].apply(_check_underscore).any()):
+                raise Exception(
+                    "Replace in 'Family ID' columns '_' on another a simbols"
+                )
+
+            _lgen["fid"] = data[fid_col]
+
+        else:
+            _lgen["fid"] = 1
+
+        _lgen["snp_name"] = data[snp_name]
+        _lgen[["allele1", "allele2"]] = data[alleles].replace({'-': 0})
+
+    except Exception as e:
+        raise e
+
+    return _lgen[_fields].astype(_f_dtype)
+
+
+def _check_underscore(value: str) -> bool:
+    """ Checking for underscore in a string
+
+    :param value: String for checked
+    :return: Return True if there is an underline in the string, False if not
+    """
+    _under_l = re.compile(r"_")
+
+    if _under_l.findall(value):
+        return True
+
+    return False

format/_snp.py

@@ -0,0 +1,113 @@
+# !/usr/bin/env python
+# coding: utf-8
+
+__author__ = "Igor Loschinin (igor.loschinin@gmail.com)"
+
+from pathlib import Path
+from .__settings import FIELDS_ILLUMIN, MAP_FIELDS
+
+import pandas as pd
+
+
+class Snp(object):
+	""" The process of converting genomic map data - FinalReport.txt obtained
+	from Illumin. Recoding allele data into quantitative data, saving in the
+	format necessary for calculating gblup on blupf90.
+
+	:argument fmt: Data format to use snp in plink and blupf90. Default
+		value "uga". """
+
+	_ALLELE_CODE = {
+		'AA': 0, 'AB': 1, 'BA': 1, 'BB': 2, '--': 5
+	}
+
+	_FIELDS = ['SNP_NAME', 'SAMPLE_ID', 'SNP']
+	_F_DTYPE = dict(zip(_FIELDS, (str for _ in range(len(_FIELDS)))))
+
+	def __init__(self, fmt: str | None = "uga") -> None:
+		self._format_data = fmt
+		self.__data_snp = None
+
+	@property
+	def data(self) -> pd.DataFrame | None:
+		return self.__data_snp
+
+	def process(self, data: pd.DataFrame) -> None:
+		""" Data processing and formatting. Calculation of statistical
+		information
+
+		:param data: Data from FinalReport file. Example:
+			SNP Name  Sample ID  Allele1 - AB  Allele2 - AB  GC Score  GT Score
+			ABCA12	14814	A	A	0.4048	0.8164
+			ARS-BFGL-BAC-13031	14814	B	B	0.9083	0.8712
+			ARS-BFGL-BAC-13039	14814	A	A	0.9005	0.9096
+			ARS-BFGL-BAC-13049	14814	A	B	0.9295	0.8926
+
+		:return: Returns true if the data was formatted successfully and
+			statistical information was calculated, false if an error.
+		"""
+
+		if not all(list(map(lambda x: x in data.columns, FIELDS_ILLUMIN))):
+			raise KeyError(
+				'The name of the fields does not match the finalreport.txt '
+				'file from Illumina'
+			)
+
+		self.__data_snp = data.rename(columns=MAP_FIELDS)
+		self.__data_snp['SNP'] = \
+			self.__data_snp[['ALLELE1', 'ALLELE2']].\
+			sum(axis=1).\
+			map(Snp._ALLELE_CODE)
+
+		self.__data_snp = self.__data_snp[Snp._FIELDS].astype(Snp._F_DTYPE)
+
+		if self._format_data is not None and self._format_data == "uga":
+			self.__data_snp = self._format_uga(
+				self.__data_snp[['SAMPLE_ID', 'SNP']]
+			)
+
+	@staticmethod
+	def _format_uga(data: pd.DataFrame) -> pd.DataFrame:
+		""" Data format to use snp in plink and blupf90. """
+
+		return data.groupby(by='SAMPLE_ID').sum().reset_index()
+
+	def to_file(self, file_path: str | Path) -> None:
+		""" Saving data to a file.
+
+		:param file_path: Path to file
+		"""
+
+		if isinstance(file_path, str):
+			file_path = Path(file_path)
+
+		if self._format_data is not None and self._format_data == "uga":
+
+			max_len = self.__data_snp["SAMPLE_ID"].str.len().max()
+
+			self.__data_snp.\
+				apply(
+					lambda x: " ".join([
+						self._add_space(x.iloc[0], max_len), x.iloc[1]
+					]),
+					axis=1
+				).\
+				to_csv(file_path, index=False, header=False)
+
+			self.__data_snp["SAMPLE_ID"] = \
+				self.__data_snp["SAMPLE_ID"].str.strip()
+
+			return None
+
+		self.__data_snp.to_csv(file_path, sep=" ", index=False)
+
+	@staticmethod
+	def _add_space(value: str, max_len: int) -> str:
+		""" Adding spaces up to the maximum length of the value in the
+		sample_id data.
+
+		:param value: Sample_id value
+		:param max_len: Max len sample_id value
+		:return: Return replacing value
+		"""
+		return "".join([value, " " * (max_len - len(value))])

format/tests/__init__.py

@@ -0,0 +1,7 @@
+#!/usr/bin/env python
+# coding: utf-8
+__author__ = "Igor Loschinin (igor.loschinin@gmail.com)"
+
+from pathlib import Path
+
+DIR_FILES = Path(__file__).parent.joinpath("files")

format/tests/test_plink_fam.py

@@ -0,0 +1,121 @@
+#!/usr/bin/env python
+# coding: utf-8
+__author__ = "Igor Loschinin (igor.loschinin@gmail.com)"
+
+from . import DIR_FILES
+from .. import make_fam
+
+import pytest
+import pandas as pd
+
+
+@pytest.fixture
+def data_fam(request) -> pd.DataFrame | None:
+	return pd.read_pickle(DIR_FILES / f"fplink/fam/{request.param}")
+
+
+class TestPlinkFormatPed(object):
+
+	@pytest.mark.parametrize("data_fam", ["file.pl"], indirect=True)
+	def test_fam_true(self, data_fam: pd.DataFrame) -> None:
+		assert not make_fam(
+			data_fam,
+			"SAMPLE_ID",
+			"SAMPLE_ID"
+		).empty
+
+		assert not make_fam(
+			data_fam,
+			"SAMPLE_ID",
+			"SAMPLE_ID"
+		).empty
+
+	def test_fam_empty(self) -> None:
+		assert make_fam(
+			pd.DataFrame(columns=["SAMPLE_ID", "SNP"]),
+			"SAMPLE_ID",
+		).empty
+
+		assert make_fam(
+			pd.DataFrame(columns=["SAMPLE_ID", "SNP"]),
+			"SAMPLE_ID",
+			"SAMPLE_ID",
+		).empty
+
+	@pytest.mark.parametrize("data_fam", ["file.pl"], indirect=True)
+	def test_fam_raise_columns(self, data_fam: pd.DataFrame) -> None:
+		# SID_COL
+		with pytest.raises(
+				KeyError, match="Data has not in name columns SAMPLE_ID1!"
+		):
+			make_fam(
+				data_fam,
+				"SAMPLE_ID1",
+				"SAMPLE_ID",
+			)
+
+		# FID_COL
+		with pytest.raises(
+				KeyError, match="Data has not in name columns SAMPLE_ID1!"
+		):
+			make_fam(
+				data_fam,
+				"SAMPLE_ID",
+				"SAMPLE_ID1"
+			)
+
+	@pytest.mark.parametrize("data_fam", ["file2.pl"], indirect=True)
+	def test_fam_raises_underscope_sid(self, data_fam: pd.DataFrame) -> None:
+
+		# SID_COL
+		with pytest.raises(
+				Exception,
+				match="Replace in 'Sample ID' columns '_' on another a simbols"
+		):
+			make_fam(
+				data_fam,
+				"SAMPLE_ID",
+				"SAMPLE_ID"
+			)
+
+	@pytest.mark.parametrize("data_fam", ["file3.pl"], indirect=True)
+	def test_fam_raises_underscope_fid(self, data_fam: pd.DataFrame) -> None:
+
+		# FID_COL
+		with pytest.raises(
+				Exception,
+				match="Replace in 'Family ID' columns '_' on another a simbols"
+		):
+			make_fam(
+				data_fam,
+				"SAMPLE_ID",
+				"FAMILY_ID"
+			)
+
+	@pytest.mark.parametrize("data_fam", ["file4.pl"], indirect=True)
+	def test_fam_check_data(self, data_fam: pd.DataFrame) -> None:
+		res = make_fam(
+			data_fam,
+			"SAMPLE_ID",
+			"FAMILY_ID",
+			father_col="father",
+			mother_col="mother",
+			sex_col="sex",
+			pheno_col="pheno"
+		)
+
+		res2 = make_fam(
+			data_fam,
+			"SAMPLE_ID",
+			"FAMILY_ID",
+		)
+
+		assert all(res.father.values == list('1234'))
+		assert all(res.mother.values == list('5678'))
+		assert all(res.sex.values == list('1210'))
+		assert all(res.pheno.values == ['12', '13', '14', '15'])
+
+		assert all(res2.father.values == list('0000'))
+		assert all(res2.mother.values == list('0000'))
+		assert all(res2.sex.values == list('0000'))
+		assert all(res2.pheno.values == ['-9', '-9', '-9', '-9'])

format/tests/test_plink_lgen.py

@@ -0,0 +1,106 @@
+#!/usr/bin/env python
+# coding: utf-8
+__author__ = "Igor Loschinin (igor.loschinin@gmail.com)"
+
+from . import DIR_FILES
+from .. import make_lgen
+
+import pytest
+import pandas as pd
+
+
+@pytest.fixture
+def data_lgen(request) -> pd.DataFrame:
+	return pd.read_pickle(DIR_FILES / f"fplink/lgen/{request.param}")
+
+
+class TestPlinkFormatLgen(object):
+
+	@pytest.mark.parametrize("data_lgen", ["file.pl"], indirect=True)
+	def test_lgen_true(self, data_lgen: pd.DataFrame) -> None:
+		assert not make_lgen(
+			data_lgen,
+			"Sample ID",
+			"SNP Name",
+			["Allele1 - AB", "Allele2 - AB"]
+		).empty
+
+	def test_lgen_empty(self) -> None:
+		assert make_lgen(
+			pd.DataFrame(columns=[
+				"Sample ID", "SNP Name", "Allele1 - AB", "Allele2 - AB"
+			]),
+			"Sample ID",
+			"SNP Name",
+			["Allele1 - AB", "Allele2 - AB"]
+		).empty
+
+	@pytest.mark.parametrize("data_lgen", ["file.pl"], indirect=True)
+	def test_lgen_raise_columns(self, data_lgen: pd.DataFrame) -> None:
+
+		with pytest.raises(
+			Exception,
+			match="Replace in 'Sample ID' columns '_' on another a simbols"
+		):
+			res1 = data_lgen.copy(deep=True)
+			res1["Sample ID"] = res1["Sample ID"] + "_"
+
+			make_lgen(
+				res1,
+				"Sample ID",
+				"SNP Name",
+				["Allele1 - AB", "Allele2 - AB"]
+			)
+
+		with pytest.raises(
+			Exception,
+			match="Replace in 'Family ID' columns '_' on another a simbols"
+		):
+			res1 = data_lgen.copy(deep=True)
+			res1["Family ID"] = res1["Sample ID"] + "_"
+
+			make_lgen(
+				res1,
+				"Sample ID",
+				"SNP Name",
+				["Allele1 - AB", "Allele2 - AB"],
+				fid_col="Family ID"
+			)
+
+		# SID
+		with pytest.raises(KeyError):
+			make_lgen(
+				data_lgen,
+				"Sample ID1",
+				"SNP Name",
+				["Allele1 - AB", "Allele2 - AB"],
+				fid_col="Family ID"
+			)
+
+		# FID_COL
+		with pytest.raises(KeyError):
+			make_lgen(
+				data_lgen,
+				"Sample ID",
+				"SNP Name",
+				["Allele1 - AB", "Allele2 - AB"],
+				fid_col="Family ID"
+			)
+
+		# SNP name
+		with pytest.raises(KeyError):
+			make_lgen(
+				data_lgen,
+				"Sample ID",
+				"SNP Name1",
+				["Allele1 - AB", "Allele2 - AB"]
+			)
+
+		# Alleles
+		with pytest.raises(KeyError):
+			make_lgen(
+				data_lgen,
+				"Sample ID",
+				"SNP Name",
+				["Allele1 - AB1", "Allele2 - AB1"]
+			)

format/tests/test_plink_map.py

@@ -0,0 +1,42 @@
+#!/usr/bin/env python
+# coding: utf-8
+__author__ = "Igor Loschinin (igor.loschinin@gmail.com)"
+
+from . import DIR_FILES
+from .. import make_map
+
+import pytest
+import pandas as pd
+
+
+@pytest.fixture
+def data_map() -> pd.DataFrame:
+	return pd.read_csv(DIR_FILES / "fplink/map/file_bovinesnp50.csv")
+
+
+class TestPlinkFormatMap(object):
+
+	def test_map_true(self, data_map) -> None:
+
+		res = make_map(data_map)
+		assert not res.empty
+
+	def test_map_raise(self, data_map) -> None:
+		with pytest.raises(
+				KeyError, match="Manifest has no data to build map format!"
+		):
+			make_map(data_map)
+			make_map(pd.DataFrame())
+			make_map(
+				pd.DataFrame(columns=['Chr', 'Name', 'MapInfo', 'morgans'])
+			)
+
+		with pytest.raises(
+				KeyError, match="Manifest has no data to build map format!"
+		):
+			make_map(pd.DataFrame())
+
+	def test_map_empty(self) -> None:
+		assert make_map(
+				pd.DataFrame(columns=['Chr', 'Name', 'MapInfo', 'morgans'])
+			).empty