CRISPRessoShared.py

'''
CRISPResso2 - Kendell Clement and Luca Pinello 2020
Software pipeline for the analysis of genome editing outcomes from deep sequencing data
(c) 2020 The General Hospital Corporation. All Rights Reserved.
'''

import argparse
import datetime
import errno
import gzip
import json
import numpy as np
import os
import pandas as pd
import re
import string
import shutil
import signal
import subprocess as sb
import unicodedata

from CRISPResso2 import CRISPResso2Align
from CRISPResso2 import CRISPRessoCOREResources

__version__ = "2.2.9"

###EXCEPTIONS############################
class FlashException(Exception):
    pass

class TrimmomaticException(Exception):
    pass

class NoReadsAlignedException(Exception):
    pass

class AlignmentException(Exception):
    pass

class SgRNASequenceException(Exception):
    pass

class NTException(Exception):
    pass

class ExonSequenceException(Exception):
    pass

class DuplicateSequenceIdException(Exception):
    pass

class NoReadsAfterQualityFilteringException(Exception):
    pass

class BadParameterException(Exception):
    pass

class AutoException(Exception):
    pass

class OutputFolderIncompleteException(Exception):
    pass

class InstallationException(Exception):
    pass

#########################################


def getCRISPRessoArgParser(parserTitle = "CRISPResso Parameters",requiredParams={}):
    parser = argparse.ArgumentParser(description=parserTitle, formatter_class=argparse.ArgumentDefaultsHelpFormatter)
    parser.add_argument('--version', action='version', version='%(prog)s '+__version__)
    parser.add_argument('-r1', '--fastq_r1', type=str,  help='First fastq file', default='', required='fastq_r1' in requiredParams)
    parser.add_argument('-r2', '--fastq_r2', type=str,  help='Second fastq file for paired end reads', default='')

    parser.add_argument('-a', '--amplicon_seq', type=str,  help='Amplicon Sequence (can be comma-separated list of multiple sequences)', required='amplicon_seq' in requiredParams)

    parser.add_argument('-an', '--amplicon_name', type=str,  help='Amplicon Name (can be comma-separated list of multiple names, corresponding to amplicon sequences given in --amplicon_seq', default='Reference')
    parser.add_argument('-amas', '--amplicon_min_alignment_score', type=str,  help='Amplicon Minimum Alignment Score; score between 0 and 100; sequences must have at least this homology score with the amplicon to be aligned (can be comma-separated list of multiple scores, corresponding to amplicon sequences given in --amplicon_seq)', default="")
    parser.add_argument('--default_min_aln_score', '--min_identity_score',  type=int, help='Default minimum homology score for a read to align to a reference amplicon', default=60)
    parser.add_argument('--expand_ambiguous_alignments', help='If more than one reference amplicon is given, reads that align to multiple reference amplicons will count equally toward each amplicon. Default behavior is to exclude ambiguous alignments.', action='store_true')
    parser.add_argument('--assign_ambiguous_alignments_to_first_reference', help='If more than one reference amplicon is given, ambiguous reads that align with the same score to multiple amplicons will be assigned to the first amplicon. Default behavior is to exclude ambiguous alignments.', action='store_true')
    parser.add_argument('-g', '--guide_seq', '--sgRNA', help="sgRNA sequence, if more than one, please separate by commas. Note that the sgRNA needs to be input as the guide RNA sequence (usually 20 nt) immediately adjacent to but not including the PAM sequence (5' of NGG for SpCas9). If the PAM is found on the opposite strand with respect to the Amplicon Sequence, ensure the sgRNA sequence is also found on the opposite strand. The CRISPResso convention is to depict the expected cleavage position using the value of the parameter '--quantification_window_center' nucleotides from the 3' end of the guide. In addition, the use of alternate nucleases besides SpCas9 is supported. For example, if using the Cpf1 system, enter the sequence (usually 20 nt) immediately 3' of the PAM sequence and explicitly set the '--cleavage_offset' parameter to 1, since the default setting of -3 is suitable only for SpCas9.", default='')
    parser.add_argument('-gn', '--guide_name', help="sgRNA names, if more than one, please separate by commas.", default='')
    parser.add_argument('-fg', '--flexiguide_seq', help="sgRNA sequence (flexible) (can be comma-separated list of multiple flexiguides). The flexiguide sequence will be aligned to the amplicon sequence(s), as long as the guide sequence has homology as set by --flexiguide_homology.")
    parser.add_argument('-fh', '--flexiguide_homology', type=int, help="flexiguides will yield guides in amplicons with at least this homology to the flexiguide sequence.", default=80)
    parser.add_argument('-fgn', '--flexiguide_name', help="flexiguide name", default='')
    parser.add_argument('--discard_guide_positions_overhanging_amplicon_edge', help="If set, for guides that align to multiple positions, guide positions will be discarded if plotting around those regions would included bp that extend beyond the end of the amplicon. ", action='store_true')
    parser.add_argument('-e', '--expected_hdr_amplicon_seq', help='Amplicon sequence expected after HDR', default='')
    parser.add_argument('-c', '--coding_seq',  help='Subsequence/s of the amplicon sequence covering one or more coding sequences for frameshift analysis. If more than one (for example, split by intron/s), please separate by commas.', default='')

    #quality filtering options
    parser.add_argument('-q', '--min_average_read_quality', type=int, help='Minimum average quality score (phred33) to keep a read', default=0)
    parser.add_argument('-s', '--min_single_bp_quality', type=int, help='Minimum single bp score (phred33) to keep a read', default=0)
    parser.add_argument('--min_bp_quality_or_N', type=int, help='Bases with a quality score (phred33) less than this value will be set to "N"', default=0)

    #output options
    parser.add_argument('--file_prefix',  help='File prefix for output plots and tables', default='')
    parser.add_argument('-n', '--name',  help='Output name of the report (default: the name is obtained from the filename of the fastq file/s used in input)', default='')
    parser.add_argument('-o', '--output_folder',  help='Output folder to use for the analysis (default: current folder)', default='')

    ## read preprocessing params
    parser.add_argument('--split_interleaved_input', '--split_paired_end', help='Splits a single fastq file containing paired end reads into two files before running CRISPResso', action='store_true')
    parser.add_argument('--trim_sequences', help='Enable the trimming of Illumina adapters with Trimmomatic', action='store_true')
    parser.add_argument('--trimmomatic_command', type=str, help='Command to run trimmomatic', default='trimmomatic')
    parser.add_argument('--trimmomatic_options_string', type=str, help='Override options for Trimmomatic, e.g. "ILLUMINACLIP:/data/NexteraPE-PE.fa:0:90:10:0:true"', default='')
    parser.add_argument('--flash_command', type=str, help='Command to run flash', default='flash')
    parser.add_argument('--min_paired_end_reads_overlap',  type=int, help='Parameter for the FLASH read merging step. Minimum required overlap length between two reads to provide a confident overlap. ', default=10)
    parser.add_argument('--max_paired_end_reads_overlap',  type=int, help='Parameter for the FLASH merging step.  Maximum overlap length expected in approximately 90%% of read pairs. Please see the FLASH manual for more information.', default=100)
    parser.add_argument('--stringent_flash_merging', help='Use stringent parameters for flash merging. In the case where flash could merge R1 and R2 reads ambiguously, the expected overlap is calculated as 2*average_read_length - amplicon_length. The flash parameters for --min-overlap and --max-overlap will be set to prefer merged reads with length within 10bp of the expected overlap. These values override the --min_paired_end_reads_overlap or --max_paired_end_reads_overlap CRISPResso parameters.', action='store_true')
    parser.add_argument('--force_merge_pairs', help=argparse.SUPPRESS, action='store_true')#help=Force-merges R1 and R2 if they cannot be merged using flash (use with caution -- may create non-biological apparent indels at the joining site)

    #quantification window params
    parser.add_argument('-w', '--quantification_window_size', '--window_around_sgrna', type=str, help='Defines the size (in bp) of the quantification window extending from the position specified by the "--cleavage_offset" or "--quantification_window_center" parameter in relation to the provided guide RNA sequence(s) (--sgRNA). Mutations within this number of bp from the quantification window center are used in classifying reads as modified or unmodified. A value of 0 disables this window and indels in the entire amplicon are considered. Default is 1, 1bp on each side of the cleavage position for a total length of 2bp. Multiple quantification window sizes (corresponding to each guide specified by --guide_seq) can be specified with a comma-separated list.', default='1')
    parser.add_argument('-wc', '--quantification_window_center', '--cleavage_offset', type=str, help="Center of quantification window to use within respect to the 3' end of the provided sgRNA sequence. Remember that the sgRNA sequence must be entered without the PAM. For cleaving nucleases, this is the predicted cleavage position. The default is -3 and is suitable for the Cas9 system. For alternate nucleases, other cleavage offsets may be appropriate, for example, if using Cpf1 this parameter would be set to 1. For base editors, this could be set to -17 to only include mutations near the 5' end of the sgRNA. Multiple quantification window centers (corresponding to each guide specified by --guide_seq) can be specified with a comma-separated list.", default='-3')
    #    parser.add_argument('--cleavage_offset', type=str, help="Predicted cleavage position for cleaving nucleases with respect to the 3' end of the provided sgRNA sequence. Remember that the sgRNA sequence must be entered without the PAM. The default value of -3 is suitable for the Cas9 system. For alternate nucleases, other cleavage offsets may be appropriate, for example, if using Cpf1 this parameter would be set to 1. To suppress the cleavage offset, enter 'N'.", default=-3)
    parser.add_argument('--exclude_bp_from_left', type=int, help='Exclude bp from the left side of the amplicon sequence for the quantification of the indels', default=15)
    parser.add_argument('--exclude_bp_from_right', type=int, help='Exclude bp from the right side of the amplicon sequence for the quantification of the indels', default=15)
    parser.add_argument('--use_legacy_insertion_quantification', help='If set, the legacy insertion quantification method will be used (i.e. with a 1bp quantification window, indels at the cut site and 1bp away from the cut site would be quantified). By default (if this parameter is not set) with a 1bp quantification window, only insertions at the cut site will be quantified.', action='store_true')

    parser.add_argument('--ignore_substitutions', help='Ignore substitutions events for the quantification and visualization', action='store_true')
    parser.add_argument('--ignore_insertions', help='Ignore insertions events for the quantification and visualization', action='store_true')
    parser.add_argument('--ignore_deletions', help='Ignore deletions events for the quantification and visualization', action='store_true')
    parser.add_argument('--discard_indel_reads', help='Discard reads with indels in the quantification window from analysis', action='store_true')

    # alignment parameters
    parser.add_argument('--needleman_wunsch_gap_open', type=int, help='Gap open option for Needleman-Wunsch alignment', default=-20)
    parser.add_argument('--needleman_wunsch_gap_extend', type=int, help='Gap extend option for Needleman-Wunsch alignment', default=-2)
    parser.add_argument('--needleman_wunsch_gap_incentive', type=int, help='Gap incentive value for inserting indels at cut sites', default=1)
    parser.add_argument('--needleman_wunsch_aln_matrix_loc', type=str, help='Location of the matrix specifying substitution scores in the NCBI format (see ftp://ftp.ncbi.nih.gov/blast/matrices/)', default='EDNAFULL')
    parser.add_argument('--aln_seed_count', type=int, default=5, help=argparse.SUPPRESS)#help='Number of seeds to test whether read is forward or reverse',default=5)
    parser.add_argument('--aln_seed_len', type=int, default=10, help=argparse.SUPPRESS)#help='Length of seeds to test whether read is forward or reverse',default=10)
    parser.add_argument('--aln_seed_min', type=int, default=2, help=argparse.SUPPRESS)#help='number of seeds that must match to call the read forward/reverse',default=2)

    #plotting parameters
    parser.add_argument('--plot_histogram_outliers', help="If set, all values will be shown on histograms. By default (if unset), histogram ranges are limited to plotting data within the 99 percentile.", action='store_true')

    #allele plot parameters
    parser.add_argument('--plot_window_size', '--offset_around_cut_to_plot',  type=int, help='Defines the size of the window extending from the quantification window center to plot. Nucleotides within plot_window_size of the quantification_window_center for each guide are plotted.', default=20)
    parser.add_argument('--min_frequency_alleles_around_cut_to_plot', type=float, help='Minimum %% reads required to report an allele in the alleles table plot.', default=0.2)
    parser.add_argument('--expand_allele_plots_by_quantification', help='If set, alleles with different modifications in the quantification window (but not necessarily in the plotting window (e.g. for another sgRNA)) are plotted on separate lines, even though they may have the same apparent sequence. To force the allele plot and the allele table to be the same, set this parameter. If unset, all alleles with the same sequence will be collapsed into one row.', action='store_true')
    parser.add_argument('--allele_plot_pcts_only_for_assigned_reference', help='If set, in the allele plots, the percentages will show the percentage as a percent of reads aligned to the assigned reference. Default behavior is to show percentage as a percent of all reads.', action='store_true')
    parser.add_argument('-qwc', '--quantification_window_coordinates', type=str, help='Bp positions in the amplicon sequence specifying the quantification window. This parameter overrides values of the "--quantification_window_center", "--cleavage_offset", "--window_around_sgrna" or "--window_around_sgrna" values. Any indels/substitutions outside this window are excluded. Indexes are 0-based, meaning that the first nucleotide is position 0. Ranges are separted by the dash sign (e.g. "start-stop"), and multiple ranges can be separated by the underscore (_). ' +
        'A value of 0 disables this filter. (can be comma-separated list of values, corresponding to amplicon sequences given in --amplicon_seq e.g. 5-10,5-10_20-30 would specify the 5th-10th bp in the first reference and the 5th-10th and 20th-30th bp in the second reference)', default=None)
    parser.add_argument('--annotate_wildtype_allele', type=str, help='Wildtype alleles in the allele table plots will be marked with this string (e.g. **).', default='')

    #output parameters
    parser.add_argument('--keep_intermediate', help='Keep all the  intermediate files', action='store_true')
    parser.add_argument('--dump', help='Dump numpy arrays and pandas dataframes to file for debugging purposes', action='store_true')
    parser.add_argument('--write_detailed_allele_table', help='If set, a detailed allele table will be written including alignment scores for each read sequence.', action='store_true')
    parser.add_argument('--fastq_output', help='If set, a fastq file with annotations for each read will be produced.', action='store_true')
    parser.add_argument('--bam_output', help='If set, a bam file with alignments for each read will be produced.', action='store_true')
    parser.add_argument('-x', '--bowtie2_index', type=str, help='Basename of Bowtie2 index for the reference genome', default='')

    #report style parameters
    parser.add_argument('--max_rows_alleles_around_cut_to_plot',  type=int, help='Maximum number of rows to report in the alleles table plot.', default=50)
    parser.add_argument('--suppress_report',  help='Suppress output report', action='store_true')
    parser.add_argument('--place_report_in_output_folder',  help='If true, report will be written inside the CRISPResso output folder. By default, the report will be written one directory up from the report output.', action='store_true')
    parser.add_argument('--suppress_plots',  help='Suppress output plots', action='store_true')
    parser.add_argument('--write_cleaned_report', action='store_true', help=argparse.SUPPRESS)#trims working directories from output in report (for web access)

    #base editor parameters
    parser.add_argument('--base_editor_output', help='Outputs plots and tables to aid in analysis of base editor studies.', action='store_true')
    parser.add_argument('--conversion_nuc_from',  help='For base editor plots, this is the nucleotide targeted by the base editor', default='C')
    parser.add_argument('--conversion_nuc_to',  help='For base editor plots, this is the nucleotide produced by the base editor', default='T')

    #prime editing parameters
    parser.add_argument('--prime_editing_pegRNA_spacer_seq', type=str, help="pegRNA spacer sgRNA sequence used in prime editing. The spacer should not include the PAM sequence. The sequence should be given in the RNA 5'->3' order, so for Cas9, the PAM would be on the right side of the given sequence.", default='')
    parser.add_argument('--prime_editing_pegRNA_extension_seq', type=str, help="Extension sequence used in prime editing. The sequence should be given in the RNA 5'->3' order, such that the sequence starts with the RT template including the edit, followed by the Primer-binding site (PBS).", default='')
    parser.add_argument('--prime_editing_pegRNA_extension_quantification_window_size', type=int, help="Quantification window size (in bp) at flap site for measuring modifications anchored at the right side of the extension sequence. Similar to the --quantification_window parameter, the total length of the quantification window will be 2x this parameter. Default: 5bp (10bp total window size)", default=5)
    parser.add_argument('--prime_editing_pegRNA_scaffold_seq', type=str, help="If given, reads containing any of this scaffold sequence before extension sequence (provided by --prime_editing_extension_seq) will be classified as 'Scaffold-incorporated'. The sequence should be given in the 5'->3' order such that the RT template directly follows this sequence. A common value is 'GGCACCGAGUCGGUGC'.", default='')
    parser.add_argument('--prime_editing_pegRNA_scaffold_min_match_length', type=int, help="Minimum number of bases matching scaffold sequence for the read to be counted as 'Scaffold-incorporated'. If the scaffold sequence matches the reference sequence at the incorporation site, the minimum number of bases to match will be minimally increased (beyond this parameter) to disambiguate between prime-edited and scaffold-incorporated sequences.", default=1)
    parser.add_argument('--prime_editing_nicking_guide_seq', type=str, help="Nicking sgRNA sequence used in prime editing. The sgRNA should not include the PAM sequence. The sequence should be given in the RNA 5'->3' order, so for Cas9, the PAM would be on the right side of the sequence", default='')
    parser.add_argument('--prime_editing_override_prime_edited_ref_seq', type=str, help="If given, this sequence will be used as the prime-edited reference sequence. This may be useful if the prime-edited reference sequence has large indels or the algorithm cannot otherwise infer the correct reference sequence.", default='')

    #special running modes
    parser.add_argument('--crispresso1_mode', help='Parameter usage as in CRISPResso 1', action='store_true')
    parser.add_argument('--dsODN', help='Label reads with the dsODN sequence provided', default='')
    parser.add_argument('--auto', help='Infer amplicon sequence from most common reads', action='store_true')
    parser.add_argument('--debug', help='Show debug messages', action='store_true')
    parser.add_argument('--no_rerun', help="Don't rerun CRISPResso2 if a run using the same parameters has already been finished.", action='store_true')
    parser.add_argument('-p', '--n_processes', type=str, help='Specify the number of processes to use for analysis.\
    Please use with caution since increasing this parameter will significantly increase the memory required to run CRISPResso. Can be set to \'max\'.', default='1')

    #processing of aligned bam files
    parser.add_argument('--bam_input', type=str,  help='Aligned reads for processing in bam format', default='')
    parser.add_argument('--bam_chr_loc', type=str,  help='Chromosome location in bam for reads to process. For example: "chr1:50-100" or "chrX".', default='')

    #deprecated params
    parser.add_argument('--save_also_png', default=False, help=argparse.SUPPRESS) #help='Save also .png images in addition to .pdf files') #depreciated -- now pngs are automatically created. Pngs can be suppressed by '--suppress_report'

    return parser

def get_crispresso_options():
    parser = getCRISPRessoArgParser(parserTitle = "Temp Params", requiredParams={})
    crispresso_options = set()
    d = parser.__dict__['_option_string_actions']
    for key in d.keys():
        d2 = d[key].__dict__['dest']
        crispresso_options.add(d2)

    return crispresso_options

def get_crispresso_options_lookup():
##dict to lookup abbreviated params
#    crispresso_options_lookup = {
#    'r1':'fastq_r1',
#    'r2':'fastq_r2',
#    'a':'amplicon_seq',
#    'an':'amplicon_name',
#    .....
#}
    crispresso_options_lookup = {}
    parser = getCRISPRessoArgParser(parserTitle = "Temp Params", requiredParams={})
    d = parser.__dict__['_option_string_actions']
    for key in d.keys():
        d2 = d[key].__dict__['dest']
        key_sub = re.sub("^-*", "", key)
        if key_sub != d2:
            crispresso_options_lookup[key_sub] = d2
    return crispresso_options_lookup


def propagate_crispresso_options(cmd, options, params):
####
# cmd - the command to run
# options - list of options to propagate e.g. crispresso options
# params - arguments given to this program

    for option in options :
        if option:
            if option in params:
                val = getattr(params, option)
                if val is None:
                    pass
                elif str(val) == "True":
                    cmd+=' --%s' % option
                elif str(val) =="False":
                    pass
                elif isinstance(val, str):
                    if val != "":
                        if re.match(r'-\d+$', val):
                            cmd+=' --%s %s' % (option, str(val))
                        elif " " in val or "-" in val:
                            cmd+=' --%s "%s"' % (option, str(val)) # quotes for options with spaces
                        else:
                            cmd+=' --%s %s' % (option, str(val))
                elif isinstance(val, bool):
                    if val:
                        cmd+=' --%s' % option
                else:
                    cmd+=' --%s %s' % (option, str(val))
    return cmd

#######
# Sequence functions
#######
nt_complement=dict({'A':'T','C':'G','G':'C','T':'A','N':'N','_':'_','-':'-'})
def reverse_complement(seq):
        return "".join([nt_complement[c] for c in seq.upper()[-1::-1]])

def reverse(seq):
    return "".join(c for c in seq.upper()[-1::-1])

def find_wrong_nt(sequence):
    return list(set(sequence.upper()).difference({'A', 'T', 'C', 'G', 'N'}))

def capitalize_sequence(x):
    return str(x).upper() if not pd.isnull(x) else x

def slugify(value): #adapted from the Django project

    value = unicodedata.normalize('NFKD', value).encode('ascii', 'ignore')
    value = re.sub(rb'[^\w\s-]', b'_', value).strip()
    value = re.sub(rb'[-\s]+', b'-', value)

    return value.decode('utf-8')


CIGAR_LOOKUP = {
    ('A', 'A'): 'M', ('A', 'C'): 'M', ('A', 'T'): 'M', ('A', 'G'): 'M',
    ('C', 'A'): 'M', ('C', 'C'): 'M', ('C', 'T'): 'M', ('C', 'G'): 'M',
    ('T', 'A'): 'M', ('T', 'C'): 'M', ('T', 'T'): 'M', ('T', 'G'): 'M',
    ('G', 'A'): 'M', ('G', 'C'): 'M', ('G', 'T'): 'M', ('G', 'G'): 'M',
    ('A', '-'): 'I', ('T', '-'): 'I', ('C', '-'): 'I', ('G', '-'): 'I',
    ('-', 'A'): 'D', ('-', 'T'): 'D', ('-', 'C'): 'D', ('-', 'G'): 'D',
    }

cigarUnexplodePattern = re.compile(r'((\w)\2{0,})')


def unexplode_cigar(exploded_cigar_string):
    """Make a CIGAR string from an exploded cigar string.
    Exploded cigar: IIMMMMMMM
    cigar_els: ['2I','7M']
    CIGAR: 2I7M

    Parameters
    ----------
    exploded_cigar_string : str
        Exploded cigar string, e.g. IIMMMMMMM

    Returns
    -------
    cigar_els : list
        List of CIGAR elements


    """
    cigar_els = []
    for (cigar_str, cigar_char) in re.findall(cigarUnexplodePattern, exploded_cigar_string):
        cigar_els.append(str(len(cigar_str)) + cigar_char)
    return cigar_els


def get_ref_length_from_cigar(cigar_string):
    """
    Given a CIGAR string, return the number of bases consumed from the
    reference sequence.
    """
    read_consuming_ops = ("M", "D", "N", "=", "X")
    result = 0
    ops = re.findall(r'(\d+)(\w)', cigar_string)
    for c in ops:
        length, op = c
        if op in read_consuming_ops:
            result += int(length)
    return result


######
# File functions
######

def clean_filename(filename):
    #get a clean name that we can use for a filename
    #validFilenameChars = "+-_.() %s%s" % (string.ascii_letters, string.digits)
    filename = str(filename).replace(' ', '_')
    validFilenameChars = "_.%s%s" % (string.ascii_letters, string.digits)

    cleanedFilename = unicodedata.normalize('NFKD', filename)
    return ''.join(c for c in cleanedFilename if c in validFilenameChars)

def check_file(filename):
    try:
        with open(filename): pass
    except IOError:
        files_in_curr_dir = os.listdir('.')
        if len(files_in_curr_dir) > 15:
            files_in_curr_dir = files_in_curr_dir[0:15]
            files_in_curr_dir.append("(Complete listing truncated)")
        dir_string = ""
        file_dir = os.path.dirname(filename)
        if file_dir == "":
            dir_string = ""
        elif os.path.isdir(file_dir):
            files_in_file_dir = os.listdir(file_dir)
            if len(files_in_file_dir) > 15:
                files_in_file_dir = files_in_file_dir[0:15]
                files_in_file_dir.append("(Complete listing truncated)")
            dir_string = "\nAvailable files in " + file_dir + ":\n\t" + "\n\t".join(files_in_file_dir)
        else:
            dir_string = "\nAdditionally, the folder '" + os.path.dirname(filename) + "' does not exist"

        raise BadParameterException("The specified file '"+filename + "' cannot be opened.\nAvailable files in current directory:\n\t" + "\n\t".join(files_in_curr_dir) + dir_string)

def force_symlink(src, dst):

    if os.path.exists(dst) and os.path.samefile(src, dst):
        return

    try:
        os.symlink(src, dst)
    except OSError as exc:
        if exc.errno == errno.EEXIST:
            os.remove(dst)
            os.symlink(src, dst)
        elif exc.errno == errno.EPROTO:
            #in docker on windows 7, symlinks don't work so well, so we'll just copy the file.
            shutil.copyfile(src, dst)

def parse_count_file(fileName):
    if os.path.exists(fileName):
        with open(fileName) as infile:
            lines = infile.readlines()
            ampSeq = lines[0].rstrip().split("\t")
            ampSeq.pop(0) #get rid of 'Amplicon' at the beginning of line
            ampSeq = "".join(ampSeq)
            lab_freqs={}
            for i in range(1, len(lines)):
                line = lines[i].rstrip()
                lab_freq_arr = line.split()
                lab = lab_freq_arr.pop(0)
                lab_freqs[lab] = lab_freq_arr
        return ampSeq, lab_freqs
    else:
        print("Cannot find output file '%s'"%fileName)
        return None, None

def parse_alignment_file(fileName):
    if os.path.exists(fileName):
        with open(fileName) as infile:
            lines = infile.readlines()
            ampSeq = lines[0].rstrip().split("\t")
            ampSeq.pop(0) #get rid of 'Amplicon' at the beginning of line
            ampSeq = "".join(ampSeq)
            lab_freqs={}
            for i in range(1, len(lines)):
                line = lines[i].rstrip()
                lab_freq_arr = line.split()
                lab = lab_freq_arr.pop(0)
                lab_freqs[lab] = lab_freq_arr
        return ampSeq, lab_freqs
    else:
        print("Cannot find output file '%s'"%fileName)
        return None, None

def check_output_folder(output_folder):
    """
    Checks to see that the CRISPResso run has completed, and gathers the amplicon info for that run
    returns:
    - quantification file = CRISPResso_quantification_of_editing_frequency.txt for this run
    - amplicons = a list of amplicons analyzed in this run
    - amplicon_info = a dict of attributes found in quantification_file for each amplicon
    """
    run_file = os.path.join(output_folder, 'CRISPResso2_info.json')
    if not os.path.exists(run_file):
        raise OutputFolderIncompleteException('The folder %s is not a valid CRISPResso2 output folder. Cannot find summary file %s.' % (output_folder, run_file))
    with open(run_file) as fh:
        run_data = json.load(fh)

    amplicon_info = {}
    amplicons = run_data['results']['ref_names']

    quantification_file = os.path.join(output_folder, run_data['running_info']['quant_of_editing_freq_filename'])
    if os.path.exists(quantification_file):
        with open(quantification_file) as quant_file:
            head_line = quant_file.readline()
            head_line_els = head_line.split("\t")
            for line in quant_file:
                line_els = line.split("\t")
                amplicon_name = line_els[0]
                amplicon_info[amplicon_name] = {}
                amplicon_quant_file = os.path.join(output_folder, run_data['results']['refs'][amplicon_name]['combined_pct_vector_filename'])
                if not os.path.exists(amplicon_quant_file):
                    raise OutputFolderIncompleteException('The folder %s is not a valid CRISPResso2 output folder. Cannot find quantification file %s for amplicon %s.' % (output_folder, amplicon_quant_file, amplicon_name))
                amplicon_info[amplicon_name]['quantification_file'] = amplicon_quant_file

                amplicon_mod_count_file = os.path.join(output_folder, run_data['results']['refs'][amplicon_name]['quant_window_mod_count_filename'])
                if not os.path.exists(amplicon_mod_count_file):
                    raise OutputFolderIncompleteException('The folder %s  is not a valid CRISPResso2 output folder. Cannot find modification count vector file %s for amplicon %s.' % (output_folder, amplicon_mod_count_file, amplicon_name))
                amplicon_info[amplicon_name]['modification_count_file'] = amplicon_mod_count_file

                if 'allele_frequency_files' in run_data['results']['refs'][amplicon_name]:
                    amplicon_info[amplicon_name]['allele_files'] = [os.path.join(output_folder, x) for x in run_data['results']['refs'][amplicon_name]['allele_frequency_files']]
                else:
                    amplicon_info[amplicon_name]['allele_files'] = []

                for idx, el in enumerate(head_line_els):
                    amplicon_info[amplicon_name][el] = line_els[idx]

        return quantification_file, amplicons, amplicon_info
    else:
        raise OutputFolderIncompleteException("The folder %s  is not a valid CRISPResso2 output folder. Cannot find quantification file '%s'." %(output_folder, quantification_file))


# Thanks https://gist.github.com/simonw/7000493 for this idea
class CRISPRessoJSONEncoder(json.JSONEncoder):
    def default(self, obj):
        if isinstance(obj, np.ndarray):
            return {
                '_type': 'np.ndarray',
                'value': obj.tolist(),
            }
        if isinstance(obj, np.integer):
            return int(obj)
        if isinstance(obj, np.floating):
            return float(obj)
        if isinstance(obj, pd.DataFrame):
            return {
                '_type': 'pd.DataFrame',
                'value': obj.to_json(orient='split'),
            }
        if isinstance(obj, datetime.datetime):
            return {
                '_type': 'datetime.datetime',
                'value': str(obj),
            }
        if isinstance(obj, datetime.timedelta):
            return {
                '_type': 'datetime.timedelta',
                'value': {
                    'days': obj.days,
                    'seconds': obj.seconds,
                    'microseconds': obj.microseconds,
                },
            }
        if isinstance(obj, set):
            return {
                '_type': 'set',
                'value': repr(obj),
            }
        if isinstance(obj, range):
            return {
                '_type': 'range',
                'value': repr(obj),
            }
        if isinstance(obj, argparse.Namespace):
            return {
                '_type': 'argparse.Namespace',
                'value': vars(obj),
            }
        return json.JSONEncoder.default(self, obj)


class CRISPRessoJSONDecoder(json.JSONDecoder):
    def __init__(self, *args, **kwargs):
        json.JSONDecoder.__init__(
            self,
            object_hook=self.object_hook,
            *args,
            **kwargs,
        )

    def object_hook(self, obj):
        if '_type' in obj:
            if obj['_type'] == 'np.ndarray':
                return np.array(obj['value'])
            if obj['_type'] == 'pd.DataFrame':
                return pd.read_json(obj['value'], orient='split')
            if obj['_type'] == 'datetime.datetime':
                return datetime.datetime.fromisoformat(obj['value'])
            if obj['_type'] == 'datetime.timedelta':
                return datetime.timedelta(
                    days=obj['value']['days'],
                    seconds=obj['value']['seconds'],
                    microseconds=obj['value']['microseconds'],
                )
            if obj['_type'] == 'set':
                return eval(obj['value'])
            if obj['_type'] == 'range':
                start, end, step = re.match(
                    r'range\((\d+), (\d+)(?:, (\d+))?\)', obj['value'],
                ).groups()
                if step is not None:
                    return range(int(start), int(end), int(step))
                return range(int(start), int(end))
            if obj['_type'] == 'argparse.Namespace':
                return argparse.Namespace(**obj['value'])
        return obj


def load_crispresso_info(
    crispresso_output_folder="",
    crispresso_info_file_name='CRISPResso2_info.json',
):
    """Load the CRISPResso2 info for a CRISPResso run.

    Parameters
    ----------
    crispresso_output_folder : string
        Path to CRISPResso folder

    Returns
    -------
    dict
        Dict of relevant information for a CRISPResso run

    """
    crispresso_info_file = os.path.join(
        crispresso_output_folder, crispresso_info_file_name,
    )
    if not os.path.isfile(crispresso_info_file):
        raise Exception('Cannot open CRISPResso info file at ' + crispresso_info_file)
    try:
        with open(crispresso_info_file) as fh:
            crispresso2_info = json.load(fh, cls=CRISPRessoJSONDecoder)
            return crispresso2_info
    except json.JSONDecodeError as e:
        raise Exception('Cannot open CRISPResso info file at ' + crispresso_info_file + "\n" + str(e))
    except (AttributeError,  EOFError, ImportError, IndexError) as e:
        # secondary errors
        raise Exception('Cannot open CRISPResso info file at ' + crispresso_info_file + "\n" + str(e))
    except Exception as e:
        raise Exception('Cannot open CRISPResso info file at ' + crispresso_info_file + "\n" + str(e))


def write_crispresso_info(crispresso_output_file, crispresso2_info):
    """Write info hash to crispresso info output file.

    Parameters
    ----------
    crispresso_output_file : string
        File path to write info to
    crispresso2_info : dict
        Dict of relevant run properties

    Returns
    -------
    Nothing

    """
    with open(crispresso_output_file, 'w') as fh:
        json.dump(crispresso2_info, fh, cls=CRISPRessoJSONEncoder)


def get_command_output(command):
    """
    Run a shell command and returns an iter to read the output.

    param:
        command: shell command to run

    returns:
        iter to read the output
    """
    p = sb.Popen(command,
                 stdout=sb.PIPE,
                 stderr=sb.STDOUT, shell=True,
                 #  encoding='utf-8',universal_newlines=True)
                 universal_newlines=True,
                 bufsize=-1)  # bufsize system default
    while(True):
        retcode = p.poll()
        line = p.stdout.readline()
        yield line
        if retcode is not None:
            break

def get_most_frequent_reads(fastq_r1, fastq_r2, number_of_reads_to_consider, flash_command, max_paired_end_reads_overlap, min_paired_end_reads_overlap, debug=False):
    """
    Get the most frequent amplicon from a fastq file (or after merging a r1 and r2 fastq file).

    Note: only works on paired end or single end reads (not interleaved)

    input:
    fastq_r1: path to fastq r1 (can be gzipped)
    fastq_r2: path to fastq r2 (can be gzipped)
    number_of_reads_to_consider: number of reads from the top of the file to examine
    min_paired_end_reads_overlap: min overlap in bp for flashing (merging) r1 and r2
    max_paired_end_reads_overlap: max overlap in bp for flashing (merging) r1 and r2

    returns:
    list of amplicon strings sorted by order in format:
    12345 AATTCCG
    124 ATATATA
    5 TTATA
    """
    view_cmd_1 = 'cat'
    if fastq_r1.endswith('.gz'):
        view_cmd_1 = 'zcat'
    file_generation_command = "%s %s | head -n %d "%(view_cmd_1, fastq_r1, number_of_reads_to_consider*4)

    if fastq_r2:
        view_cmd_2 = 'cat'
        if fastq_r2.endswith('.gz'):
            view_cmd_2 = 'zcat'
        max_overlap_param = ""
        min_overlap_param = ""
        if max_paired_end_reads_overlap:
            max_overlap_param = "--max-overlap="+str(max_paired_end_reads_overlap)
        if min_paired_end_reads_overlap:
            min_overlap_param = "--min-overlap="+str(min_paired_end_reads_overlap)
        file_generation_command = "bash -c 'paste <(%s \"%s\") <(%s \"%s\")' | head -n %d | paste - - - - | awk -v OFS=\"\\n\" -v FS=\"\\t\" '{print($1,$3,$5,$7,$2,$4,$6,$8)}' | %s - --interleaved-input --allow-outies %s %s --to-stdout 2>/dev/null " %(view_cmd_1, fastq_r1, view_cmd_2, fastq_r2, number_of_reads_to_consider*4, flash_command, max_overlap_param, min_overlap_param)
    count_frequent_cmd = file_generation_command + " | awk '((NR-2)%4==0){print $1}' | sort | uniq -c | sort -nr "
    def default_sigpipe():
        signal.signal(signal.SIGPIPE, signal.SIG_DFL)

    if (debug):
        print('command used: ' + count_frequent_cmd)

    piped_commands = count_frequent_cmd.split("|")
    pipes = [None] * len(piped_commands)
    pipes[0] = sb.Popen(piped_commands[0], stdout=sb.PIPE, preexec_fn=default_sigpipe, shell=True)
    for pipe_i in range(1, len(piped_commands)):
        pipes[pipe_i] = sb.Popen(piped_commands[pipe_i], stdin=pipes[pipe_i-1].stdout, stdout=sb.PIPE, preexec_fn=default_sigpipe, shell=True)
    top_unaligned = pipes[-1].communicate()[0]

    if pipes[-1].poll() != 0:
        raise AutoException('Cannot retrieve most frequent amplicon sequences. Got nonzero return code.')
    seq_lines = top_unaligned.decode('utf-8').strip().split("\n")
    if len(seq_lines) == 0:
        raise AutoException('Cannot parse any frequent amplicons sequences.')
    return seq_lines

def guess_amplicons(fastq_r1,fastq_r2,number_of_reads_to_consider,flash_command,max_paired_end_reads_overlap,min_paired_end_reads_overlap,aln_matrix,needleman_wunsch_gap_open,needleman_wunsch_gap_extend,min_freq_to_consider=0.2,amplicon_similarity_cutoff=0.95):
    """
    guesses the amplicons used in an experiment by examining the most frequent read (giant caveat -- most frequent read should be unmodified)
    input:
    fastq_r1: path to fastq r1 (can be gzipped)
    fastq_r2: path to fastq r2 (can be gzipped)
    number_of_reads_to_consider: number of reads from the top of the file to examine
    flash_command: command to call flash
    min_paired_end_reads_overlap: min overlap in bp for flashing (merging) r1 and r2
    max_paired_end_reads_overlap: max overlap in bp for flashing (merging) r1 and r2
    aln_matrix: matrix specifying alignment substitution scores in the NCBI format
    needleman_wunsch_gap_open: alignment penalty assignment used to determine similarity of two sequences
    needleman_wunsch_gap_extend: alignment penalty assignment used to determine similarity of two sequences
    min_freq_to_consider: selected ampilcon must be frequent at least at this percentage in the population
    amplicon_similarity_cutoff: if the current amplicon has similarity of greater than this cutoff to any other existing amplicons, it won't be added

    returns:
    list of putative amplicons
    """
    seq_lines = get_most_frequent_reads(fastq_r1, fastq_r2, number_of_reads_to_consider, flash_command, max_paired_end_reads_overlap, min_paired_end_reads_overlap)

    curr_amplicon_id = 1

    amplicon_seq_arr = []

    #add most frequent amplicon to the list
    count, seq = seq_lines[0].strip().split()
    amplicon_seq_arr.append(seq)
    curr_amplicon_id += 1

    #for the remainder of the amplicons, test them before adding
    for i in range(1, len(seq_lines)):
        count, seq = seq_lines[i].strip().split()
        last_count, last_seq = seq_lines[i-1].strip().split()
        #if this allele is present in at least XX% of the samples
#        print('debug 509 testing ' + str(seq_lines[i]) + ' with ' + str(count) + ' out of consididered ' + str(number_of_reads_to_consider) + ' min freq: ' + str(min_freq_to_consider))
        if float(last_count)/float(number_of_reads_to_consider) > min_freq_to_consider:
            this_amplicon_seq_arr = amplicon_seq_arr[:]
            this_amplicon_max_pct = 0 #keep track of similarity to most-similar already-found amplicons
            for amp_seq in this_amplicon_seq_arr:
                ref_incentive = np.zeros(len(amp_seq)+1, dtype=int)
                fws1, fws2, fwscore=CRISPResso2Align.global_align(seq, amp_seq, matrix=aln_matrix, gap_incentive=ref_incentive, gap_open=needleman_wunsch_gap_open, gap_extend=needleman_wunsch_gap_extend,)
                rvs1, rvs2, rvscore=CRISPResso2Align.global_align(reverse_complement(seq), amp_seq, matrix=aln_matrix, gap_incentive=ref_incentive, gap_open=needleman_wunsch_gap_open, gap_extend=needleman_wunsch_gap_extend,)
                #if the sequence is similar to a previously-seen read, don't add it
                min_len =  min(len(last_seq), len(seq))
                max_score = max(fwscore, rvscore)
                if max_score/float(min_len) > this_amplicon_max_pct:
                    this_amplicon_max_pct = max_score/float(min_len)
            #if this amplicon was maximally-similar to all other chosen amplicons by less than amplicon_similarity_cutoff, add to the list
            if this_amplicon_max_pct < amplicon_similarity_cutoff:
                amplicon_seq_arr.append(seq)
                curr_amplicon_id += 1
        else:
            break

    return amplicon_seq_arr

def guess_guides(amplicon_sequence,fastq_r1,fastq_r2,number_of_reads_to_consider,flash_command,max_paired_end_reads_overlap,
            min_paired_end_reads_overlap,exclude_bp_from_left,exclude_bp_from_right,
            aln_matrix,needleman_wunsch_gap_open,needleman_wunsch_gap_extend,
            min_edit_freq_to_consider=0.1,min_edit_fold_change_to_consider=3,
            pam_seq="NGG", min_pct_subs_in_base_editor_win=0.8):
    """
    guesses the guides used in an experiment by identifying the most-frequently edited positions, editing types, and PAM sites
    input:
    ampilcon_sequence - amplicon to analyze
    fastq_r1: path to fastq r1 (can be gzipped)
    fastq_r2: path to fastq r2 (can be gzipped)
    number_of_reads_to_consider: number of reads from the top of the file to examine
    flash_command: command to call flash
    min_paired_end_reads_overlap: min overlap in bp for flashing (merging) r1 and r2
    max_paired_end_reads_overlap: max overlap in bp for flashing (merging) r1 and r2
    exclude_bp_from_left: number of bp to exclude from the left side of the amplicon sequence for the quantification of the indels
    exclude_bp_from_right: number of bp to exclude from the right side of the amplicon sequence for the quantification of the indels
    aln_matrix: matrix specifying alignment substitution scores in the NCBI format
    needleman_wunsch_gap_open: alignment penalty assignment used to determine similarity of two sequences
    needleman_wunsch_gap_extend: alignment penalty assignment used to determine similarity of two sequences
    min_edit_freq_to_consider: edits must be at least this frequency for consideration
    min_edit_fold_change_to_consider: edits must be at least this fold change over background for consideration
    pam_seq: pam sequence to look for (can be regex or contain degenerate bases)
    min_pct_subs_in_base_editor_win: if at least this percent of substitutions happen in the predicted base editor window, return base editor flag

    returns:
    tuple of (putative guide, boolean is_base_editor)
    or (None, None)
    """
    seq_lines = get_most_frequent_reads(fastq_r1, fastq_r2, number_of_reads_to_consider, flash_command, max_paired_end_reads_overlap, min_paired_end_reads_overlap)

    amp_len = len(amplicon_sequence)
    gap_incentive = np.zeros(amp_len+1, dtype=int)
    include_idxs=range(amp_len)
    exclude_idxs=[]


    if exclude_bp_from_left:
       exclude_idxs+=range(exclude_bp_from_left)
    if exclude_bp_from_right:
       exclude_idxs+=range(amp_len)[-exclude_bp_from_right:]
    include_idxs=np.ravel(include_idxs)
    exclude_idxs=np.ravel(exclude_idxs)

    include_idxs=set(np.setdiff1d(include_idxs, exclude_idxs))

    all_indel_count_vector = np.zeros(amp_len)
    all_sub_count_vector = np.zeros(amp_len)
    tot_count = 0;
    for i in range(len(seq_lines)):
        count, seq = seq_lines[i].strip().split()
        count = int(count)
        tot_count += count
        fws1, fws2, fwscore=CRISPResso2Align.global_align(seq, amplicon_sequence, matrix=aln_matrix, gap_incentive=gap_incentive,
            gap_open=needleman_wunsch_gap_open, gap_extend=needleman_wunsch_gap_extend,)
        payload=CRISPRessoCOREResources.find_indels_substitutions(fws1, fws2, include_idxs)
        all_indel_count_vector[payload['all_insertion_positions']]+=count
        all_indel_count_vector[payload['all_deletion_positions']]+=count
        all_sub_count_vector[payload['all_substitution_positions']]+=count

    background_val = np.mean(all_indel_count_vector)
    if len(exclude_idxs) > 0:
        all_indel_count_vector[exclude_idxs] = 0
    max_loc = np.argmax(all_indel_count_vector)
    max_val = all_indel_count_vector[max_loc]

    #return nothing if the max edit doesn't break threshold
    if max_val/float(tot_count) < min_edit_freq_to_consider:
        return (None, None)

    #return nothing if the max edit doesn't break threshold over background
    if max_val/background_val < min_edit_fold_change_to_consider:
        return(None, None)


    pam_regex_string = pam_seq.upper()
    pam_regex_string = pam_regex_string.replace('I', '[ATCG]')
    pam_regex_string = pam_regex_string.replace('N', '[ATCG]')
    pam_regex_string = pam_regex_string.replace('R', '[AG]')
    pam_regex_string = pam_regex_string.replace('Y', '[CT]')
    pam_regex_string = pam_regex_string.replace('S', '[GC]')
    pam_regex_string = pam_regex_string.replace('W', '[AT]')
    pam_regex_string = pam_regex_string.replace('K', '[GT]')
    pam_regex_string = pam_regex_string.replace('M', '[AC]')
    pam_regex_string = pam_regex_string.replace('B', '[CGT]')
    pam_regex_string = pam_regex_string.replace('D', '[AGT]')
    pam_regex_string = pam_regex_string.replace('H', '[ACT]')
    pam_regex_string = pam_regex_string.replace('V', '[ACG]')

    is_base_editor = False
    #offset from expected position
    for offset in (0, +1, -1, +2, +3, +4, -2):
        #forward direction
        #find pam near max edit loc
        pam_start = max_loc+4 + offset
        pam_end = max_loc+7 + offset
        guide_start = max_loc-16 + offset
        guide_end = max_loc+4 + offset
        base_edit_start = max_loc-16 + offset
        base_edit_end = max_loc-6 + offset
        if pam_start > 0 and guide_end < amp_len:
            if re.match(pam_regex_string, amplicon_sequence[pam_start:pam_end]):
                guide_seq = amplicon_sequence[guide_start:guide_end]
                sum_base_edits = sum(all_sub_count_vector[base_edit_start:base_edit_end])
                #if a lot of edits are in the predicted base editor window, set base editor true
                #specifically, if at least min_pct_subs_in_base_editor_win % of substitutions happen in the predicted base editor window
                if sum_base_edits > min_pct_subs_in_base_editor_win * sum(all_sub_count_vector):
                    is_base_editor = True
                return(guide_seq, is_base_editor)

        #reverse direction
        pam_start = max_loc-5 - offset
        pam_end = max_loc-2 - offset
        guide_start = max_loc-2 - offset
        guide_end = max_loc+18 - offset
        base_edit_start = max_loc+8 - offset
        base_edit_end = max_loc+18 - offset
        if pam_start > 0 and guide_end < amp_len:
            if re.match(pam_regex_string, amplicon_sequence[pam_start:pam_end]):
                guide_seq = amplicon_sequence[guide_start:guide_end]
                sum_base_edits = sum(all_sub_count_vector[base_edit_start:base_edit_end])
                #if a lot of edits are in the predicted base editor window, set base editor true
                #specifically, if at least min_pct_subs_in_base_editor_win % of substitutions happen in the predicted base editor window
                if sum_base_edits > min_pct_subs_in_base_editor_win * sum(all_sub_count_vector):
                    is_base_editor = True
                return(guide_seq, is_base_editor)

    return (None, None)


######
# Fastq file manipulation
######

def force_merge_pairs(r1_filename, r2_filename, output_filename):
    """
    This can be useful in case paired end reads are too short to cover the amplicon.
    Note that this should be used with extreme caution because non-biological indels will appear at the site of read merging.
    R1------>     <-------R2
    becomes
    R1------><------R2

    input:
    r1_filename: path to fastq r1 (can be gzipped)
    r2_filename: path to fastq r2 (can be gzipped)
    output_filename: path to merged output filename

    returns:
    linecount: the number of lines of the resulting file
    """

    if r1_filename.endswith('.gz'):
        f1 = gzip.open(r1_filename, 'rt')
    else:
        f1 = open(r1_filename, 'r')
    if r2_filename.endswith('.gz'):
        f2 = gzip.open(r2_filename, 'rt')
    else:
        f2 = open(r2_filename, 'r')

    if output_filename.endswith('.gz'):
        f_out = gzip.open(output_filename, 'wt')
    else:
        f_out = open(output_filename, 'w')

    lineCount = 0
    id1 = f1.readline()
    while id1 :
        lineCount += 1
        seq1 = f1.readline()
        seq1 = seq1.strip()
        plus1 = f1.readline()
        qual1 = f1.readline()
        qual1 = qual1.strip()

        id2 = f2.readline()
        seq2 = reverse_complement(f2.readline().strip())+"\n"
        plus2 = f2.readline()
        qual2 = f2.readline()

        f_out.write(id1+seq1+seq2+plus1+qual1+qual2)

        id1 = f1.readline()
    f1.close()
    f2.close()
    f_out.close()

    return(lineCount)

######
# allele modification functions
######

def get_row_around_cut(row, cut_point, offset):
    cut_idx=row['ref_positions'].index(cut_point)
    return row['Aligned_Sequence'][cut_idx-offset+1:cut_idx+offset+1], row['Reference_Sequence'][cut_idx-offset+1:cut_idx+offset+1], row['Read_Status']=='UNMODIFIED', row['n_deleted'], row['n_inserted'], row['n_mutated'], row['#Reads'], row['%Reads']


def get_dataframe_around_cut(df_alleles, cut_point,offset,collapse_by_sequence=True):
    if df_alleles.shape[0] == 0:
        return df_alleles
    ref1 = df_alleles['Reference_Sequence'].iloc[0]
    ref1 = ref1.replace('-', '')
    if (cut_point + offset + 1 > len(ref1)):
        raise(BadParameterException('The plotting window cannot extend past the end of the amplicon. Amplicon length is ' + str(len(ref1)) + ' but plot extends to ' + str(cut_point+offset+1)))

    df_alleles_around_cut=pd.DataFrame(list(df_alleles.apply(lambda row: get_row_around_cut(row, cut_point, offset), axis=1).values),
                    columns=['Aligned_Sequence', 'Reference_Sequence', 'Unedited', 'n_deleted', 'n_inserted', 'n_mutated', '#Reads', '%Reads'])

    df_alleles_around_cut=df_alleles_around_cut.groupby(['Aligned_Sequence', 'Reference_Sequence', 'Unedited', 'n_deleted', 'n_inserted', 'n_mutated']).sum().reset_index().set_index('Aligned_Sequence')

    df_alleles_around_cut.sort_values(by='%Reads', inplace=True, ascending=False)
    df_alleles_around_cut['Unedited']=df_alleles_around_cut['Unedited']>0
    return df_alleles_around_cut

def get_row_around_cut_debug(row, cut_point, offset):
    cut_idx=row['ref_positions'].index(cut_point)
    #don't check overflow -- it was checked when program started
    return row['Aligned_Sequence'][cut_idx-offset+1:cut_idx+offset+1], row['Reference_Sequence'][cut_idx-offset+1:cut_idx+offset+1], row['Read_Status']=='UNMODIFIED', row['n_deleted'], row['n_inserted'], row['n_mutated'], row['#Reads'], row['%Reads'], row['Aligned_Sequence'], row['Reference_Sequence']

def get_dataframe_around_cut_debug(df_alleles, cut_point, offset):
    df_alleles_around_cut=pd.DataFrame(list(df_alleles.apply(lambda row: get_row_around_cut_debug(row, cut_point, offset), axis=1).values),
                    columns=['Aligned_Sequence', 'Reference_Sequence', 'Unedited', 'n_deleted', 'n_inserted', 'n_mutated', '#Reads', '%Reads', 'oSeq', 'oRef'])
    df_alleles_around_cut=df_alleles_around_cut.groupby(['Aligned_Sequence', 'Reference_Sequence', 'Unedited', 'n_deleted', 'n_inserted', 'n_mutated', 'oSeq', 'oRef']).sum().reset_index().set_index('Aligned_Sequence')

    df_alleles_around_cut.sort_values(by='%Reads', inplace=True, ascending=False)
    df_alleles_around_cut['Unedited']=df_alleles_around_cut['Unedited']>0
    return df_alleles_around_cut

def get_amplicon_info_for_guides(ref_seq,guides,guide_mismatches,guide_names,quantification_window_centers,quantification_window_sizes,quantification_window_coordinates,exclude_bp_from_left,exclude_bp_from_right,plot_window_size,guide_plot_cut_points,discard_guide_positions_overhanging_amplicon_edge=False):
    """
    gets cut site and other info for a reference sequence and a given list of guides

    input:
    ref_seq : reference sequence
    guides : a list of guide sequences
    guide_mismatches : a list of positions where a guide may have mismatches (for flexiguides)
    guide_names : a list of names for each guide
    quantification_window_centers : a list of positions where quantification is centered for each guide
    quantification_window_sizes : a list of lengths of quantification windows extending from quantification_window_center for each guide
    quantification_window_coordinates: if given, these override quantification_window_center and quantification_window_size for setting quantification window. These are specific for this amplicon.
    exclude_bp_from_left : these bp are excluded from the quantification window
    exclude_bp_from_right : these bp are excluded from the quantification window
    plot_window_size : length of window extending from quantification_window_center to plot
    guide_plot_cut_points : whether or not to add cut point to plot (prime editing flaps don't have cut points)
    discard_guide_positions_overhanging_amplicon_edge : if True, for guides that align to multiple positions, guide positions will be discarded if plotting around those regions would included bp that extend beyond the end of the amplicon.

    returns:
    this_sgRNA_sequences : list of sgRNAs that are in this amplicon
    this_sgRNA_intervals : indices of each guide
    this_sgRNA_cut_points : cut points for each guide (defined by quantification_window_center)
    this_sgRNA_plot_cut_points : whether or not a cut point is plotted
    this_sgRNA_plot_idxs : list of indices to be plotted for each sgRNA
    this_sgRNA_mismatches: list of mismatches between the guide and the amplicon
    this_sgRNA_names : list of names for each sgRNA (to disambiguate in case a sequence aligns to multiple positions)
    this_include_idxs : list of indices to be included in quantification
    this_exclude_idxs : list of indices to be excluded from quantification
    """
    ref_seq_length = len(ref_seq)

    this_sgRNA_sequences = []
    this_sgRNA_intervals = []
    this_sgRNA_cut_points = []
    this_sgRNA_plot_cut_points = []
    this_sgRNA_plot_idxs=[]
    this_sgRNA_mismatches = []
    this_sgRNA_names = []
    this_include_idxs=[]
    this_exclude_idxs=[]

    if exclude_bp_from_left:
       this_exclude_idxs+=range(exclude_bp_from_left)

    if exclude_bp_from_right:
       this_exclude_idxs+=range(ref_seq_length)[-exclude_bp_from_right:]

    window_around_cut=max(1, plot_window_size)

    seen_cut_points = {} #keep track of cut points in case 2 gudes cut at same position (so they can get different names)
    seen_guide_names = {} #keep track of guide names (so we don't assign a guide the same name as another guide)
    for guide_idx, current_guide_seq in enumerate(guides):
        if current_guide_seq == '':
            continue
        offset_fw=quantification_window_centers[guide_idx]+len(current_guide_seq)-1
        offset_rc=(-quantification_window_centers[guide_idx])-1

        #.. run once with findall to get number of matches
        fw_matches = re.findall(current_guide_seq, ref_seq, flags=re.IGNORECASE)
        rv_matches = re.findall(reverse_complement(current_guide_seq), ref_seq, flags=re.IGNORECASE)
        match_count = len(fw_matches) + len(rv_matches)

        #and now create the iter which will keep track of the locations of matches
        fw_matches = re.finditer(current_guide_seq, ref_seq, flags=re.IGNORECASE)
        rv_matches = re.finditer(reverse_complement(current_guide_seq), ref_seq, flags=re.IGNORECASE)

        #for every match, append:
        # this_sgRNA_cut_points, this_sgRNA_intervals,this_sgRNA_mismatches,this_sgRNA_names,this_sgRNA_sequences,include_idxs
        for m in fw_matches:
            cut_p = m.start() + offset_fw
            if discard_guide_positions_overhanging_amplicon_edge and ((cut_p - window_around_cut + 1 < 0) or (cut_p + window_around_cut > ref_seq_length-1)):
                continue
            this_sgRNA_cut_points.append(cut_p)
            this_sgRNA_plot_cut_points.append(guide_plot_cut_points[guide_idx])
            this_sgRNA_intervals.append((m.start(), m.start() + len(current_guide_seq)-1))
            this_sgRNA_mismatches.append(guide_mismatches[guide_idx])

            if quantification_window_sizes[guide_idx] > 0:
                st=max(0, cut_p-quantification_window_sizes[guide_idx]+1)
                en=min(ref_seq_length-1, cut_p+quantification_window_sizes[guide_idx]+1)
                this_include_idxs.extend(range(st, en))

            this_sgRNA_name = guide_names[guide_idx]
            if match_count == 1:
                this_sgRNA_names.append(this_sgRNA_name)
            else:
                if this_sgRNA_name == "":
                    this_sgRNA_name = current_guide_seq.upper()
                this_potential_name = this_sgRNA_name + "_" + str(m.start())
                curr_guide_idx = 1
                while this_potential_name in seen_guide_names:
                    curr_guide_idx += 1
                    this_potential_name = this_sgRNA_name + "_" + str(m.start()) + "_" + curr_guide_idx
                this_sgRNA_names.append(this_potential_name)
            this_sgRNA_sequences.append(current_guide_seq.upper())

        for m in rv_matches:
            cut_p = m.start() + offset_rc #cut position
            if discard_guide_positions_overhanging_amplicon_edge and ((cut_p - window_around_cut + 1 < 0) or (cut_p + window_around_cut > ref_seq_length-1)):
                continue
            this_sgRNA_cut_points.append(cut_p)
            this_sgRNA_plot_cut_points.append(guide_plot_cut_points[guide_idx])
            this_sgRNA_intervals.append((m.start(), m.start() + len(current_guide_seq)-1))
            this_sgRNA_mismatches.append([len(current_guide_seq)-(x+1) for x in guide_mismatches[guide_idx]])

            if quantification_window_sizes[guide_idx] > 0:
                st=max(0, cut_p-quantification_window_sizes[guide_idx]+1)
                en=min(ref_seq_length-1, cut_p+quantification_window_sizes[guide_idx]+1)
                this_include_idxs.extend(range(st, en))

            this_sgRNA_name = guide_names[guide_idx]
            if match_count == 1:
                this_sgRNA_names.append(this_sgRNA_name)
            else:
                if this_sgRNA_name == "":
                    this_sgRNA_name = current_guide_seq.upper()
                this_potential_name = this_sgRNA_name + "_" + str(m.start())
                curr_guide_idx = 1
                while this_potential_name in seen_guide_names:
                    curr_guide_idx += 1
                    this_potential_name = this_sgRNA_name + "_" + str(m.start()) + "_" + curr_guide_idx
                this_sgRNA_names.append(this_potential_name)
            this_sgRNA_sequences.append(current_guide_seq.upper())


    #create mask of positions in which to include/exclude indels for the quantification window
    #first, if exact coordinates have been given, set those
    given_include_idxs = []
    if quantification_window_coordinates is not None:
        coordinate_include_idxs = []
        theseCoords = str(quantification_window_coordinates).split("_")
        for coord in theseCoords:
            coordRE = re.match(r'^(\d+)-(\d+)$', coord)
            if coordRE:
                start = int(coordRE.group(1))
                end = int(coordRE.group(2)) + 1
                if end > ref_seq_length:
                    raise NTException("End coordinate " + str(end) + " for '" + str(coord) + "' in '" + str(theseCoords) + "' is longer than the sequence length ("+str(ref_seq_length)+")")
                coordinate_include_idxs.extend(range(start, end))
            else:
                raise NTException("Cannot parse analysis window coordinate '" + str(coord) + "' in '" + str(theseCoords) + "'. Coordinates must be given in the form start-end e.g. 5-10 . Please check the --analysis_window_coordinate parameter.")
        given_include_idxs = coordinate_include_idxs
        this_include_idxs = coordinate_include_idxs
    #otherwise, take quantification centers + windows calculated for each guide above
    elif this_sgRNA_cut_points and len(this_include_idxs) > 0:
        given_include_idxs = this_include_idxs
    else:
        this_include_idxs=range(ref_seq_length)

    #flatten the arrays to avoid errors with old numpy library
    this_include_idxs=np.ravel(this_include_idxs)
    this_exclude_idxs=np.ravel(this_exclude_idxs)
    given_include_idxs=np.ravel(given_include_idxs)
    pre_exclude_include_idxs = this_include_idxs.copy()

    this_include_idxs=set(np.setdiff1d(this_include_idxs, this_exclude_idxs))
    if len(np.setdiff1d(given_include_idxs, list(this_include_idxs))) > 0:
        raise BadParameterException('The quantification window has been partially exluded by the --exclude_bp_from_left or --exclude_bp_from_right parameters. Given: ' + str(given_include_idxs) + ' Pre: ' + str(pre_exclude_include_idxs) + ' Post: ' + str(this_include_idxs))
    if len(this_include_idxs) == 0:
        if len(pre_exclude_include_idxs) > 0:
            raise BadParameterException('The quantification window around the sgRNA is excluded. Please decrease the exclude_bp_from_right and exclude_bp_from_left parameters.')
        else:
            raise BadParameterException('The entire sequence has been excluded. Please enter a longer amplicon, or decrease the exclude_bp_from_right and exclude_bp_from_left parameters')

    if this_sgRNA_cut_points and plot_window_size>0:
        for cut_p in this_sgRNA_cut_points:
            if cut_p - window_around_cut + 1 < 0:
                raise BadParameterException('Offset around cut would extend to the left of the amplicon. Please decrease plot_window_size parameter. Cut point: ' + str(cut_p) + ' window: ' + str(window_around_cut) + ' reference: ' + str(ref_seq_length))
            if cut_p + window_around_cut > ref_seq_length-1:
                raise BadParameterException('Offset around cut would be greater than reference sequence length. Please decrease plot_window_size parameter. Cut point: ' + str(cut_p) + ' window: ' + str(window_around_cut) + ' reference: ' + str(ref_seq_length))
            st=max(0, cut_p-window_around_cut+1)
            en=min(ref_seq_length-1, cut_p+window_around_cut+1)
            this_sgRNA_plot_idxs.append(sorted(list(range(st, en))))
    else:
       this_sgRNA_plot_idxs.append(range(ref_seq_length))

    this_include_idxs = np.sort(list(this_include_idxs))
    this_exclude_idxs = np.sort(list(this_exclude_idxs))

    return this_sgRNA_sequences, this_sgRNA_intervals, this_sgRNA_cut_points, this_sgRNA_plot_cut_points, this_sgRNA_plot_idxs, this_sgRNA_mismatches, this_sgRNA_names, this_include_idxs, this_exclude_idxs

def set_guide_array(vals, guides, property_name):
    """
    creates an int array of vals of the same length as guide, filling in missing values with the first item in vals
    vals: comma-separated string of values
    guides: list of guides
    property_name: property name, useful for creating warning to user

    returns: list of int vals
    """
    #if only one is given, set it for all guides
    vals_array = str(vals).split(",")
    ret_array = [int(vals_array[0])]*len(guides)
    #len(vals_array) is always one -- don't freak out if it's longer than 0 guides
    if len(vals_array) > 1 and len(vals_array) > len(guides):
        raise BadParameterException("More %s values were given than guides. Guides: %d %ss: %d"%(property_name, len(guides), property_name, len(vals_array)))

    if len(guides) == 0:
        return []

    for idx, val in enumerate(vals_array):
        if val != '':
            ret_array[idx] = int(val)
    return ret_array



def get_alignment_coordinates(to_sequence,from_sequence,aln_matrix,needleman_wunsch_gap_open,needleman_wunsch_gap_extend):
    """
    gets the coordinates to align from from_sequence to to_sequence
    such that from_sequence[i] matches to to_sequence[inds[i]]
    params:
    to_sequence : sequence to align to
    from_sequence: sequence to align from
    aln_matrix: matrix specifying alignment substitution scores in the NCBI format
    needleman_wunsch_gap_open: needleman wunsch gap open penalty
    needleman_wunsch_gap_extend: needleman wunsch gap extend penalty

    returns:
    inds_l : if there's a gap in to_sequence, these values are filled with the left value
    inds_r : if there's a gap in to_sequence, these values are filled with the right value (after the gap)
    """
    this_gap_incentive = np.zeros(len(from_sequence)+1, dtype=int)
    fws1, fws2, fwscore=CRISPResso2Align.global_align(to_sequence, from_sequence, matrix=aln_matrix, gap_open=needleman_wunsch_gap_open, gap_extend=needleman_wunsch_gap_extend, gap_incentive=this_gap_incentive)
#    print(fws1)
#    print(fws2)
    s1inds_l = []
    s1inds_r = []
    s1ix_l = -1
    s1ix_r = 0
    s2ix = -1
    for ix in range(len(fws1)):
        if fws1[ix] != "-":
            s1ix_l += 1
        if fws2[ix] != "-":
            s2ix += 1
            s1inds_l.append(s1ix_l)
            s1inds_r.append(s1ix_r)
        if fws1[ix] != "-":
            s1ix_r += 1
    return s1inds_l, s1inds_r

def get_mismatches(seq_1, seq_2,aln_matrix,needleman_wunsch_gap_open,needleman_wunsch_gap_extend):
    """
    for a specific sequence seq_1, gets the location of mismatches as an array of length(seq_1) with reference to seq_2
    Only searches in the positive direction
    Only positions of seq_1 covered in seq_2 are included (e.g. if the base in seq_1 is deleted in seq_2 it won't be included)
    params:
    seq_1: sequence to compare
    seq_2: sequence to compare against
    aln_matrix: matrix specifying alignment substitution scores in the NCBI format. Can be read by: aln_matrix = CRISPResso2Align.read_matrix(aln_matrix_loc)
    needleman_wunsch_gap_open: needleman wunsch gap open penalty
    needleman_wunsch_gap_extend: needleman wunsch gap extend penalty

    returns:
    mismatches_arr: positions in seq_1 that mismatch seq_2
    """
    #when we set up the gap_flanking score, this should be set to the gap open penalty -- we'd like a good end-to-end alignment here
    coords_l, coords_r = get_alignment_coordinates(seq_2, seq_1,aln_matrix,needleman_wunsch_gap_open,needleman_wunsch_gap_extend)
    mismatch_coords = []
    for i in range(len(seq_1)):
        if coords_l[i] < 0 or coords_l[i] >= len(seq_2) or seq_1[i] != seq_2[coords_l[i]] or coords_r[i] >= len(seq_2) or seq_1[i] != seq_2[coords_r[i]]:
            mismatch_coords.append(i)
    return mismatch_coords

def get_best_aln_pos_and_mismatches(guide_seq, within_amp_seq, aln_matrix, needleman_wunsch_gap_open, needleman_wunsch_gap_extend=0):
    """
    for a specific guide, gets the location, sequence, and mismatches for that guide at that location
    Only searches in the positive direction
    params:
    guide_seq: short guide sequence to look for
    within_amp_seq: longer amplicon to search within
    aln_matrix: matrix specifying alignment substitution scores in the NCBI format
    needleman_wunsch_gap_open: gap open penalty for needleman wunsch
    needleman_wunsch_gap_extend: gap extend pentaly for needleman wunsch (set as 0 by default to allow for longer stretches of gaps)

    returns:
    best_aln_seq: sequence in within_amp_seq of best hit
    best_aln_score: score of best alignment
    best_aln_mismatches: mismatches between best_aln_seq and guide_seq
    best_aln_start: start location of best match
    best_aln_end: end location of best match (1pb after last base)
    s1: aligned s1 (guide_seq)
    s2: aligned s2 (within_amp_seq)
    """
    ref_incentive = np.zeros(len(within_amp_seq)+1, dtype=int)
    s1, s2, fw_score=CRISPResso2Align.global_align(guide_seq, within_amp_seq, matrix=aln_matrix, gap_incentive=ref_incentive, gap_open=needleman_wunsch_gap_open, gap_extend=0,)
    range_start_dashes = 0
    m = re.search(r'^-+', s1)
    if (m):
        range_start_dashes = m.span()[1]
    range_end_dashes = len(within_amp_seq)
    m = re.search(r'-+$', s1)
    if (m):
        range_end_dashes = m.span()[0]
    guide_seq_in_amp = s2[range_start_dashes:range_end_dashes].replace("-", "")

    best_aln_seq = guide_seq_in_amp
    best_aln_mismatches = get_mismatches(guide_seq, guide_seq_in_amp,aln_matrix,needleman_wunsch_gap_open,needleman_wunsch_gap_extend)
    best_aln_start = range_start_dashes
    best_aln_end = range_end_dashes
    return(best_aln_seq, fw_score, best_aln_mismatches, best_aln_start, best_aln_end, s1, s2)

def get_sgRNA_mismatch_vals(seq1, seq2, start_loc, end_loc, coords_l, coords_r, rev_coords_l, rev_coords_r):
    """
    given coordinates between one sequence and the other, given a start and end position, finds the locations of mismatches between the sequences between the start and end positions
    the sgRNA (as defined between start_loc and end_loc) is a substring of one of the samples, but the entire seqs are aligned to avoid any funny partial alignments
    seq1 : first sequence to compare
    seq2 : second sequence to compare
    start_loc : location to start comparison
    end_loc : location to stop comparison
    coords_l, coords_r : cooresponding indices between seq1 -> seq2
    rev_coords_l, rev_coords_r : cooresponding indices between seq2 -> seq1

    returns:
    mismatches between the two strings (for use as sgRNA_mismatches)
    """
    this_mismatches = []
    seen_inds = {} #detect deletions in other string
    last_mismatch_val = rev_coords_l[coords_l[start_loc]] #detect deletions in this string
    for i in range(coords_l[start_loc], coords_r[end_loc]):
        if seq1[i] != seq2[rev_coords_l[i]] or seq1[i] == "-":
            this_mismatches.append(i-coords_l[start_loc])
        this_last_mismatch_val = rev_coords_l[i]
        if this_last_mismatch_val in seen_inds:
            this_mismatches.append(i-coords_l[start_loc])
        elif this_last_mismatch_val != last_mismatch_val: #if we skipped an index (deletion in other sequence)
            this_mismatches.append(i-coords_l[start_loc]-0.5)
        seen_inds[this_last_mismatch_val] = 1
        last_mismatch_val = this_last_mismatch_val + 1
    return(list(set(this_mismatches)))

######
# terminal functions
######
def get_crispresso_logo():
    return (r'''
     _
    '  )
    .-'
   (____
C)|     \
  \     /
   \___/
''')

def get_crispresso_header(description, header_str):
    """
    Creates the CRISPResso header string with the header_str between two crispresso mugs
    """
    term_width = 80

    logo = get_crispresso_logo()
    logo_lines = logo.splitlines()
    max_logo_width = max([len(x) for x in logo_lines])

    output_line = ""
    if header_str is not None:
        header_str = header_str.rstrip()

        header_lines = header_str.splitlines()
        while len(header_lines) < len(logo_lines):
            header_lines = [""] + header_lines
        while len(header_lines) > len(logo_lines):
            logo_lines = [""] + logo_lines

        max_header_width = max([len(x) for x in header_lines])


        pad_space = int((term_width - (max_logo_width*2) - max_header_width) / 4)
        pad_string = " " * pad_space

        for i in range(len(logo_lines))[::-1]:
            output_line = (logo_lines[i].ljust(max_logo_width) + pad_string + header_lines[i].ljust(max_header_width) + pad_string + logo_lines[i].ljust(max_logo_width)).center(term_width) + "\n" + output_line

    else:
        pad_space = int((term_width - max_logo_width) / 2)
        pad_string = " " * pad_space
        for i in range(len(logo_lines))[::-1]:
            output_line = (pad_string + logo_lines[i].ljust(max_logo_width) + pad_string).center(term_width) + "\n" + output_line

    output_line += '\n'+('[CRISPResso version ' + __version__ + ']').center(term_width) + '\n' + ('[Note that starting in version 2.1.0 insertion quantification has been changed\nto only include insertions completely contained by the quantification window.\nTo use the legacy quantification method (i.e. include insertions directly adjacent\nto the quantification window) please use the parameter --use_legacy_insertion_quantification]').center(term_width) + "\n" + ('[For support contact kclement@mgh.harvard.edu]').center(term_width) + "\n"

    description_str = ""
    for str in description:
        str = str.strip()
        description_str += str.center(term_width) + "\n"

    return "\n" + description_str + output_line

def get_crispresso_footer():
    logo = get_crispresso_logo()
    logo_lines = logo.splitlines()

    max_logo_width = max([len(x) for x in logo_lines])
    pad_space = int((80 - max_logo_width) / 2)
    pad_string = " " * pad_space

    output_line = ""
    for i in range(len(logo_lines))[::-1]:
        output_line = pad_string + logo_lines[i].ljust(max_logo_width) + pad_string + "\n" + output_line

    return output_line