haphic

#!/usr/bin/env python3
# -*- coding: utf-8 -*-

# Author: Xiaofei Zeng
# Email: xiaofei_zeng@whu.edu.cn
# Created Time: 2022-09-13 22:47


"""This code is only a wrapper for HapHiC scripts"""


import os
import sys
import subprocess
try:
    from importlib.metadata import distribution
except:
    from pkg_resources import get_distribution as distribution
from scripts._version import __version__, __update_time__


TITLE_DOC = """
          \033[42;37m         ______  __              ______  _______________        \033[0m
          \033[42;37m         ___  / / /_____ ___________  / / /__(_)_  ____/        \033[0m
          \033[42;37m         __  /_/ /_  __ `/__  __ \\_  /_/ /__  /_  /             \033[0m
          \033[42;37m         _  __  / / /_/ /__  /_/ /  __  / _  / / /___           \033[0m
          \033[42;37m         /_/ /_/  \\__,_/ _  .___//_/ /_/  /_/  \\____/           \033[0m
          \033[42;37m                         /_/                                    \033[0m
          \033[42;37m             HapHiC: A fast, reference-independent,             \033[0m
          \033[42;37m        allele-aware scaffolding tool based on Hi-C data        \033[0m
          \033[47;32m               Author  :  Xiaofei Zeng                          \033[0m
          \033[47;32m               Email   :  xiaofei_zeng@whu.edu.cn               \033[0m
          \033[47;32m               Version :  {}                                 \033[0m
          \033[47;32m               Update  :  {}                            \033[0m
""".format(__version__, __update_time__)

HELP_DOC = """
  Usage: haphic [command] <parameters>

  Command    Description
  --------   ----------------------------------------------------------------------
  pipeline   Execute an automated pipeline consisting of "cluster", "reassign",
             "sort", and "build". For better result through parameter tuning
             or to parallelize the "sort" step on a computer cluster, you may
             run the commands below individually:

  cluster    [Step1] Execute preprocessing and Markov clustering for contigs.
  reassign   [Step2] Reassign and rescue contigs. Perform additional agglomerative
                     hierarchical clustering if necessary.
  sort       [Step3] Order and orient contigs within each group.
  build      [Step4] Build final scaffolds (pseudomolecules).

  plot       Draw Hi-C contact map for the whole genome or each scaffold.
  check      Check whether all dependencies are correctly installed.
  refsort    Order and orient whole scaffolds based on a reference genome.

  allhic     Call a modified version of ALLHiC, in which the hot-start optimization
             has been fixed (source code: https://github.com/zengxiaofei/allhic).
  juicer     Show the guide to generate files for Juicebox visualization.

  Use haphic [command] --help/-h to see detailed help for each individual command.

  Visit the GitHub repository for more information:
  https://github.com/zengxiaofei/HapHiC
  """

JUICE_DOC = """
  [Step 0] Install 3D-DNA, matlock, and Juicebox scripts
     (1) 3D-DNA: https://github.com/aidenlab/3d-dna
     (2) matlock: https://github.com/phasegenomics/matlock
     (3) Juicebox scripts: https://github.com/phasegenomics/juicebox_scripts

  [Step 1] Generate .mnd files
      $ /path/to/matlock bam2 juicer <filtered_bam> out.links.mnd
      $ sort -k2,2 -k6,6 out.links.mnd > out.sorted.links.mnd

  [Step 2] Generate .assembly file
      $ /path/to/juicebox_scripts/agp2assembly.py scaffolds.agp scaffolds.assembly

  [Step 3] Generate .hic file
      $ bash /path/to/3d-dna/visualize/run-assembly-visualizer.sh -p false scaffolds.assembly out.sorted.links.mnd
  """

# a dict mapping commands to corresponding programs
param_dict = {
        'cluster': 'HapHiC_cluster.py',
        'reassign': 'HapHiC_reassign.py',
        'sort': 'HapHiC_sort.py',
        'build': 'HapHiC_build.py',
        'refsort': 'HapHiC_refsort.py',
        'plot': 'HapHiC_plot.py',
        'pipeline': 'HapHiC_pipeline.py',
        'allhic': 'allhic'}


def check_packages(pkg, pkg_name):

    try:
        print('Checking {}... '.format(pkg_name), end='')
        __import__(pkg)
        print('\033[32mSuccessful (version: {}).\033[0m'.format(distribution(pkg_name).version))
    except:
        print('\033[31mFailed. {} is not correctly installed.\033[0m'.format(pkg_name))


# commands
if len(sys.argv) > 1 and sys.argv[1] in param_dict:

    script_realpath = os.path.dirname(os.path.realpath(__file__))
    sub_program = os.path.join(script_realpath, 'scripts', param_dict[sys.argv[1]])

    if len(sys.argv[2:]) == 0:
        commands = [sub_program, '-h']
    else:
        commands = [sub_program] + sys.argv[2:]

    subprocess.run(commands, check=True)

# print help message
elif len(sys.argv) == 1 or (len(sys.argv) > 1 and sys.argv[1] in {'-h', '--help'}):

    print(TITLE_DOC)
    print(HELP_DOC)


elif sys.argv[1] == 'check':
    for pkg, pkg_name in [
            ('pysam', 'pysam'), ('portion', 'portion'), 
            ('numpy', 'numpy'), ('scipy', 'scipy'), 
            ('sklearn', 'scikit-learn'), 
            ('networkx', 'networkx'), 
            ('sparse_dot_mkl', 'sparse_dot_mkl'), 
            ('matplotlib', 'matplotlib')]:

        check_packages(pkg, pkg_name)


elif sys.argv[1] in {'-v', '--version'}:
    print('HapHiC', __version__)

# for downstream analysis
elif sys.argv[1].lower() in {'juice', 'juicer', 'juicebox'}:

    print(JUICE_DOC)

# exceptions
else:

    print(TITLE_DOC)
    print(HELP_DOC)

    raise RuntimeError('Bad command or parameters! Please check the help message above.')