Source code for backbone.bipopulations

# !/usr/bin/env python3
"""Module containing the BIPopulations class and the command line interface."""

import shutil
import argparse

import matplotlib.pyplot as plt
import pandas as pd
from numpy import nan
from biobb_dna.utils.loader import read_series
from biobb_dna.utils.transform import inverse_complement
from biobb_common.generic.biobb_object import BiobbObject
from biobb_common.tools.file_utils import launchlogger
from biobb_common.tools import file_utils as fu
from biobb_common.configuration import settings


[docs]class BIPopulations(BiobbObject): """ | biobb_dna BIPopulations | Calculate BI/BII populations from epsilon and zeta parameters. Args: input_epsilC_path (str): Path to .ser file for helical parameter 'epsilC'. File type: input. `Sample file <https://raw.githubusercontent.com/bioexcel/biobb_dna/master/biobb_dna/test/data/backbone/canal_output_epsilC.ser>`_. Accepted formats: ser (edam:format_2330). input_epsilW_path (str): Path to .ser file for helical parameter 'epsilW'. File type: input. `Sample file <https://raw.githubusercontent.com/bioexcel/biobb_dna/master/biobb_dna/test/data/backbone/canal_output_epsilW.ser>`_. Accepted formats: ser (edam:format_2330). input_zetaC_path (str): Path to .ser file for helical parameter 'zetaC'. File type: input. `Sample file <https://raw.githubusercontent.com/bioexcel/biobb_dna/master/biobb_dna/test/data/backbone/canal_output_zetaC.ser>`_. Accepted formats: ser (edam:format_2330). input_zetaW_path (str): Path to .ser file for helical parameter 'zetaW'. File type: input. `Sample file <https://raw.githubusercontent.com/bioexcel/biobb_dna/master/biobb_dna/test/data/backbone/canal_output_zetaW.ser>`_. Accepted formats: ser (edam:format_2330). output_csv_path (str): Path to .csv file where output is saved. File type: output. `Sample file <https://raw.githubusercontent.com/bioexcel/biobb_dna/master/biobb_dna/test/reference/backbone/bipop_ref.csv>`_. Accepted formats: csv (edam:format_3752). output_jpg_path (str): Path to .jpg file where output is saved. File type: output. `Sample file <https://raw.githubusercontent.com/bioexcel/biobb_dna/master/biobb_dna/test/reference/backbone/bipop_ref.jpg>`_. Accepted formats: jpg (edam:format_3579). properties (dict): * **sequence** (*str*) - (None) Nucleic acid sequence corresponding to the input .ser file. Length of sequence is expected to be the same as the total number of columns in the .ser file, minus the index column (even if later on a subset of columns is selected with the *seqpos* option). * **seqpos** (*list*) - (None) list of sequence positions (columns indices starting by 0) to analyze. If not specified it will analyse the complete sequence. Examples: This is a use example of how to use the building block from Python:: from biobb_dna.backbone.bipopulations import bipopulations prop = { 'sequence': 'GCAT', } bipopulations( input_epsilC_path='/path/to/epsilC.ser', input_epsilW_path='/path/to/epsilW.ser', input_zetaC_path='/path/to/zetaC.ser', input_zetaW_path='/path/to/zetaW.ser', output_csv_path='/path/to/table/output.csv', output_jpg_path='/path/to/table/output.jpg', properties=prop) Info: * wrapped_software: * name: In house * license: Apache-2.0 * ontology: * name: EDAM * schema: http://edamontology.org/EDAM.owl """ def __init__(self, input_epsilC_path, input_epsilW_path, input_zetaC_path, input_zetaW_path, output_csv_path, output_jpg_path, properties=None, **kwargs) -> None: properties = properties or {} super().__init__(properties) # Input/Output files self.io_dict = { 'in': { 'input_epsilC_path': input_epsilC_path, 'input_epsilW_path': input_epsilW_path, 'input_zetaC_path': input_zetaC_path, 'input_zetaW_path': input_zetaW_path }, 'out': { 'output_csv_path': output_csv_path, 'output_jpg_path': output_jpg_path } } self.properties = properties self.sequence = properties.get("sequence") self.seqpos = properties.get("seqpos", None)
[docs] @launchlogger def launch(self) -> int: """Execute the :class:`BIPopulations <backbone.bipopulations.BIPopulations>` object.""" # Check the properties fu.check_properties(self, self.properties) # check sequence if self.sequence is None or len(self.sequence) < 2: raise ValueError("sequence is null or too short!") # check seqpos if self.seqpos is not None: if (max(self.seqpos) > len(self.sequence) - 2) or (min(self.seqpos) < 1): raise ValueError( f"seqpos values must be between 1 and {len(self.sequence) - 2}") if not (isinstance(self.seqpos, list) and len(self.seqpos) > 1): raise ValueError( "seqpos must be a list of at least two integers") # Creating temporary folder self.tmp_folder = fu.create_unique_dir(prefix="backbone_") fu.log('Creating %s temporary folder' % self.tmp_folder, self.out_log) # Copy input_file_path1 to temporary folder shutil.copy(self.io_dict['in']['input_epsilC_path'], self.tmp_folder) shutil.copy(self.io_dict['in']['input_epsilW_path'], self.tmp_folder) shutil.copy(self.io_dict['in']['input_zetaC_path'], self.tmp_folder) shutil.copy(self.io_dict['in']['input_zetaW_path'], self.tmp_folder) # read input files epsilC = read_series( self.io_dict['in']['input_epsilC_path'], usecols=self.seqpos) epsilW = read_series( self.io_dict['in']['input_epsilW_path'], usecols=self.seqpos) zetaC = read_series( self.io_dict['in']['input_zetaC_path'], usecols=self.seqpos) zetaW = read_series( self.io_dict['in']['input_zetaW_path'], usecols=self.seqpos) # calculate difference between epsil and zeta parameters xlabels = self.get_xlabels( self.sequence, inverse_complement(self.sequence)) diff_epsil_zeta = self.get_angles_difference( epsilC, zetaC, epsilW, zetaW) # calculate BI population BI = (diff_epsil_zeta < 0).sum(axis=0) * 100 / len(diff_epsil_zeta) BII = 100 - BI # save table Bpopulations_df = pd.DataFrame({ "Nucleotide": xlabels, "BI population": BI, "BII population": BII}) Bpopulations_df.to_csv( self.io_dict['out']['output_csv_path'], index=False) # save plot fig, axs = plt.subplots(1, 1, dpi=300, tight_layout=True) axs.bar( range(len(xlabels)), BI, label="BI") axs.bar( range(len(xlabels)), BII, bottom=BI, label="BII") # empty bar to divide both sequences axs.bar( [len(BI)//2], [100], color='white', label=None) axs.legend() axs.set_xticks(range(len(xlabels))) axs.set_xticklabels(xlabels, rotation=90) axs.set_xlabel("Nucleotide Sequence") axs.set_ylabel("BI/BII Population (%)") axs.set_title("Nucleotide parameter: BI/BII Population") fig.savefig( self.io_dict['out']['output_jpg_path'], format="jpg") plt.close() # Remove temporary file(s) if self.remove_tmp: self.tmp_files.append(self.tmp_folder) self.remove_tmp_files() return 0
[docs] def get_xlabels(self, strand1, strand2): # get list of tetramers, except first and last two bases labelsW = list(strand1) labelsW[0] = f"{labelsW[0]}5\'" labelsW[-1] = f"{labelsW[-1]}3\'" labelsW = [ f"{i}-{j}" for i, j in zip(labelsW, range(1, len(labelsW)+1))] labelsC = list(strand2)[::-1] labelsC[0] = f"{labelsC[0]}5\'" labelsC[-1] = f"{labelsC[-1]}3\'" labelsC = [ f"{i}-{j}" for i, j in zip(labelsC, range(len(labelsC), 0, -1))] if self.seqpos is not None: labelsC = [labelsC[i] for i in self.seqpos] labelsW = [labelsW[i] for i in self.seqpos] xlabels = labelsW + ['-'] + labelsC return xlabels
[docs] def get_angles_difference(self, epsilC, zetaC, epsilW, zetaW): # concatenate zeta and epsil arrays separator_df = pd.DataFrame({"-": nan}, index=range(len(zetaW))) zeta = pd.concat([ zetaW, separator_df, zetaC[zetaC.columns[::-1]]], axis=1) epsil = pd.concat([ epsilW, separator_df, epsilC[epsilC.columns[::-1]]], axis=1) # difference between epsilon and zeta coordinates diff_epsil_zeta = epsil - zeta return diff_epsil_zeta
[docs]def bipopulations( input_epsilC_path: str, input_epsilW_path: str, input_zetaC_path: str, input_zetaW_path: str, output_csv_path: str, output_jpg_path: str, properties: dict = None, **kwargs) -> int: """Create :class:`BIPopulations <dna.backbone.bipopulations.BIPopulations>` class and execute the: meth: `launch() <dna.backbone.bipopulations.BIPopulations.launch>` method. """ return BIPopulations( input_epsilC_path=input_epsilC_path, input_epsilW_path=input_epsilW_path, input_zetaC_path=input_zetaC_path, input_zetaW_path=input_zetaW_path, output_csv_path=output_csv_path, output_jpg_path=output_jpg_path, properties=properties, **kwargs).launch()
[docs]def main(): """Command line execution of this building block. Please check the command line documentation.""" parser = argparse.ArgumentParser(description='Calculate BI/BII populations.', formatter_class=lambda prog: argparse.RawTextHelpFormatter(prog, width=99999)) parser.add_argument('--config', required=False, help='Configuration file') required_args = parser.add_argument_group('required arguments') required_args.add_argument('--input_epsilC_path', required=True, help='Helical parameter <epsilC> input ser file path. Accepted formats: ser.') required_args.add_argument('--input_epsilW_path', required=True, help='Helical parameter <epsilW> input ser file path. Accepted formats: ser.') required_args.add_argument('--input_zetaC_path', required=True, help='Helical parameter <zetaC> input ser file path. Accepted formats: ser.') required_args.add_argument('--input_zetaW_path', required=True, help='Helical parameter <zetaW> input ser file path. Accepted formats: ser.') required_args.add_argument('--output_csv_path', required=True, help='Path to output csv file. Accepted formats: csv.') required_args.add_argument('--output_jpg_path', required=True, help='Path to output jpg file. Accepted formats: jpg.') args = parser.parse_args() args.config = args.config or "{}" properties = settings.ConfReader(config=args.config).get_prop_dic() bipopulations( input_epsilC_path=args.input_epsilC_path, input_epsilW_path=args.input_epsilW_path, input_zetaC_path=args.input_zetaC_path, input_zetaW_path=args.input_zetaW_path, output_csv_path=args.output_csv_path, output_jpg_path=args.output_jpg_path, properties=properties)
if __name__ == '__main__': main()