tb-profiler

#! /usr/bin/env python3
import sys
import pathogenprofiler as pp
from pathogenprofiler import TempFilePrefix, TempFolder
import pathogenprofiler.variant_calling as vc
import argparse
from rich_argparse import ArgumentDefaultsRichHelpFormatter
import json
import tbprofiler as tbp
import os
import csv
from uuid import uuid4
import glob
import atexit
import shutil
from joblib import Parallel, delayed
from tqdm import tqdm
import logging
from rich.logging import RichHandler
from copy import deepcopy

import importlib
import pkgutil


discovered_plugins = {
    name: importlib.import_module(name)
    for finder, name, ispkg
    in pkgutil.iter_modules()
    if name.startswith('tbprofiler_')
}




__softwarename__ = 'tbprofiler'

@atexit.register
def cleanup():
    if "last_traceback" in vars(sys):
        if args.files_prefix and not args.no_clean:
            sys.stderr.write("Cleaning up after failed run\n")
            for f in glob.glob(args.files_prefix+"*"):
                os.remove(f)
        import traceback
        
        if "prefix" in vars(args):
            outfile = "%s.errlog.txt" % args.prefix
        elif "vcf" in vars(args):
            outfile = "%s.errlog.txt" % args.vcf.split("/")[-1]
        else:
            outfile = "%s.errlog.txt" % uuid4() 
        if "conf" in vars(args) and args.conf:
            del args.conf['json_db']
        with open(outfile, "w") as O:
            O.write("# tb-profiler error report\n\n")
            O.write("* OS: %s\n" % sys.platform)
            O.write("* Program version: %s\n" % tbp.__version__)
            O.write("* Database version: %s\n" % args.conf.get("version","N/A") if "conf" in vars(args) and args.conf else "")
            O.write("* Program call:\n")
            O.write("```\n")
            O.write("%s\n" % vars(args))
            O.write("```\n")

            O.write("## Traceback:\n")
            O.write("```\n")
            traceback.print_tb(sys.last_traceback,file=O)
            O.write("```\n")

            O.write("## Value:\n")
            O.write("```\n")
            O.write("%s" % sys.last_value)
            O.write("```\n")
        logging.error("""\n
################################# ERROR #######################################

This run has failed. Please check all arguments and make sure all input files
exist. If no solution is found, please open up an issue at
https://github.com/jodyphelan/TBProfiler/issues/new and paste or attach the
contents of the error log (%s)

###############################################################################
""" % (outfile))

def vcf_job(args: argparse.Namespace,sample_name: str):
    # logging.info(f"\nExtracting variants and running pipeline for {sample_name}\n")
    copy_of_args = deepcopy(args)
    copy_of_args.prefix = sample_name
    with TempFolder() as tmpfolder:
        cmd = f"bcftools view -s {sample_name} -ac 1 {args.vcf} | bcftools +fixploidy -Oz -o {tmpfolder}/{args.vcf} && bcftools index {tmpfolder}/{args.vcf} " % vars(copy_of_args)
        pp.run_cmd(cmd)
        os.chdir(tmpfolder)
        copy_of_args.files_prefix = os.path.abspath(f"{os.getcwd()}/{copy_of_args.prefix}")
        main_profile(copy_of_args)
        os.chdir('../')


def multisample_vcf_run(args):
    vcf_obj = pp.Vcf(args.vcf)
    jobs = []

    for sample_name in vcf_obj.samples:
        jobs.append((args,sample_name))

    parallel = Parallel(n_jobs=args.threads, return_as='generator')
    [r for r in tqdm(parallel(delayed(vcf_job)(cmd[0],cmd[1]) for cmd in jobs),total=len(jobs),desc="Running jobs")]

def create_output_directories(args,directories=["bam","vcf","results"]):
    if pp.nofolder(args.dir):
        os.mkdir(args.dir)
    for d in directories:
        if pp.nofolder(args.dir+"/"+d):
            os.mkdir(args.dir+"/"+d)

def plugin_decorator(func):
    def wrapper(args):
        for plugin in tbp.ProfilePlugin.__subclasses__():
            logging.debug(f"Running pre process for {plugin}")
            plugin().pre_process(args)
        func(args)
        for plugin in tbp.ProfilePlugin.__subclasses__():
            logging.debug(f"Running post process for {plugin}")
            plugin().post_process(args)
    return wrapper

@plugin_decorator
def main_profile(args):
        
    pp.process_args(args)
    create_output_directories(args)

    if args.vcf:
        nsamples = len(tbp.get_vcf_samples(args.vcf))
        if nsamples>1:
            return multisample_vcf_run(args)

    tbp.process_tb_profiler_args(args)

    variants_profile = pp.run_profiler(args)

    notes = set()
    if 'rules' in args.conf:
        rules_applied = pp.apply_rules(args.conf['rules'], variants_profile, just_make_note=True if args.implement_rules==False else False)
        for var in variants_profile:
            for ann in var.annotation:
                if 'note' in ann:
                    notes.add(ann['note'])

    else:
        rules_applied = []

    logging.debug(rules_applied)
    tbp.clean_up_duplicate_annotations(variants_profile)
    # Convert variant objects to DrVariant if they cause resistance
    for var in variants_profile:
        var.convert_to_dr_element()


    if args.call_lineage:
        barcode_result = tbp.barcode2lineage(pp.run_barcoder(args))
        

    if args.spoligotype:
        spoligotype = tbp.spoligotype(args)
    else:
        spoligotype = None

    if args.data_source in ('bam','fastq'):
        qc = pp.run_bam_qc(args)
    elif args.data_source=="fasta":
        qc = pp.run_fasta_qc(args)
    else:
        qc = pp.run_vcf_qc(args)


    result = tbp.create_resistance_result(
        args = args,
        notes=notes,
        lineage=barcode_result,
        spoligotype=spoligotype,
        variants=variants_profile,
        qc=qc,
    )


    if args.snp_dist:
        tbp.run_snp_dists(args,result)
        tbp.update_neighbour_snp_dist_output(args,result)

    for plugin in tbp.ProfilePlugin.__subclasses__():
        plugin().run(args)


    ### Create folders for results if they don't exist ###
    if pp.nofolder(d:=args.dir+"/bam") and args.read1:
        os.mkdir(d)
    if pp.nofolder(d:=args.dir+"/vcf"):
        os.mkdir(d)
    if pp.nofolder(d:=args.dir+"/results"):
        os.mkdir(d)




    tbp.write_outputs(args,result)




    # ### Move files to respective directories ###
    if os.path.isfile("%s.nwk" % args.files_prefix):
        shutil.copy("%s.nwk" % args.files_prefix,"%s/results/latest.nwk" % (args.dir))

    result_files = {
    
        "%s.targets.vcf.gz" % args.files_prefix: "%s/vcf/%s.targets.vcf.gz" % (args.dir,args.prefix),
        "%s.vcf.gz" % args.files_prefix: "%s/vcf/%s.vcf.gz" % (args.dir,args.prefix),
        "%s.bam" % args.files_prefix: "%s/bam/%s.bam" % (args.dir,args.prefix),
        "%s.bam.bai" % args.files_prefix: "%s/bam/%s.bam.bai" % (args.dir,args.prefix),
        "%s.nwk" % args.files_prefix: "%s/results/%s.nwk" % (args.dir,args.prefix),
    }
    if args.save_low_dp_mask:
        result_files["%s.%s.mask.bed" % (args.files_prefix,args.prefix)] = "%s/results/%s.mask.bed" % (args.dir,args.prefix)
    for file,target in result_files.items():
        if os.path.isfile(file):
            shutil.move(file,target)

    if not args.no_clean:
        pp.run_cmd("rm %s*" % args.files_prefix)
    
    
    pp.logging.info("[green]Profiling finished sucessfully![/]",extra={"markup": True})

def main_update_tbdb(args):

    if args.match_ref:
        args.match_ref = os.path.abspath(args.match_ref)

    if pp.nofolder("tbdb"):
        pp.run_cmd(f"git clone {args.repo}")
    os.chdir("tbdb")
    pp.run_cmd(f'git checkout {args.branch}')

    pp.run_cmd("git pull")
    if args.commit:
        pp.run_cmd(f"git checkout {args.commit}")

    if args.prefix==None:
            args.prefix = args.branch


    extra_args = []
    if os.path.isfile('rules.txt'):
        extra_args.append("--rules rules.txt")
    if args.match_ref:
        extra_args.append("--match_ref %s" % os.path.abspath(args.match_ref))

    extra_args = " ".join(extra_args)
    with TempFilePrefix() as tmpfile:
        pp.run_cmd(f"tb-profiler create_db --prefix {args.prefix} --csv mutations.csv --watchlist watchlist.csv {extra_args} --load")

    os.chdir("../")
    pp.logging.info("Sucessfully updated TBDB")

def main_create_db(args):

    version_string = json.load(open('variables.json'))['tb-profiler-version']
    tbp.check_db_version(version_string,tbp.__version__)

    if args.no_overwrite:
        dbs = pp.list_db(args.software_name)
        names = [x['version']['name'] for x in dbs]
        if args.prefix in names:
            pp.warninglog("\nWarning: A database with this name already exists!\n")
            sys.exit()

    extra_files = {
        "spoligotype_spacers": args.spoligotypes,
        "spoligotype_annotations": args.spoligotype_annotations,
        "bedmask": {"name":args.bedmask,"convert":1},
    }
    if args.barcode:
        extra_files["barcode"] = args.barcode
    if args.rules:
        extra_files["rules"] = args.rules
    
    with TempFilePrefix() as tmpfile:
        args.csv = tbp.reformat_variant_csv_file(args.csv,f'{tmpfile}.variants.csv')
        args.watchlist = tbp.reformat_variant_csv_file([args.watchlist],f'{tmpfile}.watchlist.csv')

        pp.create_db(args,extra_files=extra_files)

def main_load_library(args):
    variables_file = "%(prefix)s.variables.json" % vars(args)
    source_dir = os.path.realpath(args.dir)
    pp.load_db(variables_file,args.software_name,source_dir=source_dir)
    


def main_lineage(args):
    create_output_directories(args,directories=["results"])
    args.data_source = "bam"
    barcode_result = tbp.barcode2lineage(pp.run_barcoder(args))
    result = tbp.create_lineage_result(args,barcode_result)
    json_output = args.dir+"/results/"+args.prefix+".lineage.json"
    open(json_output,"w").write(result.model_dump_json(indent=4))

def main_spoligotype(args):
    create_output_directories(args,directories=["results"])
    if args.bam:
        args.bam_file = args.bam
    spoligotype = tbp.spoligotype(args)
    json_output = args.dir+"/results/"+args.prefix+".spoligotype.json"
    open(json_output,"w").write(spoligotype.model_dump_json(indent=4))
    

def main_collate(args):
    tbp.collate_results(args)

def main_version(args):
    sys.stdout.write("tb-profiler version %s\n" % tbp.__version__)

def main_reformat(args):
    data = json.load(open(args.json))
    result = tbp.ProfileResult(**data)
    args.prefix = result.id
    tbp.write_outputs(args,result,template_file=args.text_template)

def main_batch(args):
    if args.args==None:
        args.args=""
    commands = []


    for row in csv.DictReader(open(args.csv)):
        line = f"tb-profiler profile -p {row['id']} --temp {args.temp} --dir {args.dir} --threads {args.threads_per_job}"
        if "bam" in row:
            line += f" -a {row['bam']}" 
        elif "read1" in row:
            line += f" -1 {row['read1']}"
            if "read2" in row:
                line += f" -2 {row['read2']}"
        elif "vcf" in row:
            line += f" -v {row['vcf']}"
        elif "fasta" in row:
            line += f" -f {row['fasta']}"
        cmd = line + " " + args.args 
        commands.append(cmd)

    # run commands in parallel with joblib 
    parallel = Parallel(n_jobs=args.jobs, return_as='generator')
    [r for r in tqdm(parallel(delayed(pp.run_cmd)(cmd) for cmd in commands),total=len(commands),desc="Running jobs")]
    # tqdm(Parallel(n_jobs=args.jobs, return_as='generator')(delayed(pp.run_cmd)(cmd) for cmd in commands))
    

def main_list_db(args):
    dbs = pp.list_db(args.software_name)
    for db in dbs:
        if 'version' in db:
            d = dict(**db['version'], location=f"{sys.base_prefix}/share/{args.software_name}/{db['version']['name']}")
            sys.stdout.write("%(name)s\t%(commit)s\t%(Author)s\t%(Date)s\t%(location)s\n" % d)





#### Argument Parsing ####

def int_2_or_more(arg):
    try:
        i = int(arg)
    except ValueError:    
        raise argparse.ArgumentTypeError("Must be a of type Integer")
    if i < 2 :
        raise argparse.ArgumentTypeError("Argument must be < " + "2")
    return i




variant_callers = [cls.__software__ for cls in vc.VariantCaller.__subclasses__()]
docx_plugins = [cls.__template_name__ for cls in tbp.docx.DocxResultTemplate.__subclasses__()]




parser = argparse.ArgumentParser(description='tb-profiler: a tool to predict drug resistance and infer lineages',formatter_class=ArgumentDefaultsRichHelpFormatter,add_help=False)
parser.add_argument('--version', action='version', version="tb-profiler version %s" % tbp.__version__)
parser.add_argument('--logging',type=str.upper,default="INFO",choices=["DEBUG","INFO","WARNING","ERROR","CRITICAL"],help='Logging level')
parser.add_argument('--debug',action='store_true',help=argparse.SUPPRESS)
subparsers = parser.add_subparsers(help="Task to perform")

# Profile #
parser_sub = subparsers.add_parser('profile', help='Run whole profiling pipeline', formatter_class=ArgumentDefaultsRichHelpFormatter)
input=parser_sub.add_argument_group("Input options")
group = input.add_mutually_exclusive_group(required=True)
group.add_argument('--read1','-1',help='First read file')
input.add_argument('--read2','-2',help='Second read file')
group.add_argument('--bam','-a',help='BAM file (make sure it has been generated using the H37Rv genome (GCA_000195955.2))')
group.add_argument('--fasta','-f',help='Fasta file')
group.add_argument('--vcf','-v',help='VCF file')
input.add_argument('--platform','-m',choices=["illumina","nanopore","pacbio"],default="illumina",help='NGS Platform used to generate data')
input.add_argument('--db',default='tbdb',help='Mutation panel name')
input.add_argument('--external_db','--external-db',type=str,help='Path to db files prefix (overrides --db parameter)')

output=parser_sub.add_argument_group("Output options")
output.add_argument('--prefix','-p',default="tbprofiler",help='Sample prefix for all results generated')
output.add_argument('--csv',action="store_true",help="Add CSV output")
output.add_argument('--txt',action="store_true",help="Add text output")
output.add_argument('--text_template','--text-template',type=str,help='Jinja2 formatted template for output')
output.add_argument('--docx',action="store_true",help="Add docx output")
output.add_argument('--docx_template','--docx-template',help="Supply custom template for --docx output")
output.add_argument('--docx_plugin','--docx-plugin',choices=docx_plugins,help="Use a plugin template for --docx output")
output.add_argument('--add_columns','--add-columns',default=None,type=str,help="Add additional columns found in the mutation database to the text and csv results")
output.add_argument('--add_mutation_metadata','--add-mutation-metadata',action="store_true",help=argparse.SUPPRESS)
output.add_argument('--dir','-d',default=".",help='Storage directory')

filtering=parser_sub.add_argument_group("Variant filtering options")
filtering.add_argument('--depth',default="0,10",type=str,help='Minimum depth hard and soft cutoff specified as comma separated values')
filtering.add_argument('--af',default="0,0.1",type=str,help='Minimum allele frequency hard and soft cutoff specified as comma separated values')
filtering.add_argument('--strand',default="0,3",type=str,help='Minimum read number per strand hard and soft cutoff specified as comma separated values')
filtering.add_argument('--sv_depth','--sv-depth',default="0,10",type=str,help='Structural variant minimum depth hard and soft cutoff specified as comma separated values')
filtering.add_argument('--sv_af','--sv-af',default="0.5,0.9",type=str,help='Structural variant minimum allele frequency hard cutoff specified as comma separated values')
filtering.add_argument('--sv_len','--sv-len',default="100000,50000",type=str,help='Structural variant maximum size hard and soft cutoff specified as comma separated values')

algorithm=parser_sub.add_argument_group("Algorithm options")
algorithm.add_argument('--mapper',default="bwa", choices=["bwa","minimap2","bowtie2","bwa-mem2"],help="Mapping tool to use. If you are using nanopore or pacbio data it will default to minimap2",type=str)
algorithm.add_argument('--caller',default="freebayes", choices=variant_callers,help="Variant calling tool to use.",type=str)

algorithm.add_argument('--calling_params','--calling-params',type=str,help='Override default parameters for variant calling')
for cls in vc.VariantCaller.__subclasses__():
    if hasattr(cls,"__cli_params__"):
        for a in cls.__cli_params__:
            algorithm.add_argument(*a['args'],**a['kwargs'])

algorithm.add_argument('--kmer_counter','--kmer-counter',default='kmc',choices=["kmc","dsk"],type=str,help="Kmer counter")
algorithm.add_argument('--coverage_tool','--coverage-tool',default='samtools',choices=["samtools","bedtools"],type=str,help="Kmer counter")
algorithm.add_argument('--suspect',action="store_true",help="Use the suspect suite of tools to add ML predictions")
algorithm.add_argument('--spoligotype',action="store_true",help="Perform in-silico spoligotyping")
# algorithm.add_argument('--update_phylo','--update-phylo',action="store_true",help="Update phylogeny using usher (experimental feature)")
algorithm.add_argument('--call_whole_genome','--call-whole-genome',action="store_true",help="Call variant across the whole genome")
algorithm.add_argument('--snp_dist','--snp-dist',type=int,help="Store variant set and get all samples with snp distance less than this cutoff (experimental feature)")
algorithm.add_argument('--snp_diff_db','--snp-diff_db',type=str,help=argparse.SUPPRESS)
algorithm.add_argument('--snp_diff_no_store','--snp-diff-no-store',action='store_true',help=argparse.SUPPRESS)
algorithm.add_argument('--no_trim','--no-trim',action="store_true",help="Don't trim files using trimmomatic")
algorithm.add_argument('--no_coverage_qc','--no-coverage-qc',action="store_true",help="Don't collect flagstats")
algorithm.add_argument('--no_samclip','--no-samclip',action="store_true",help="Don't remove clipped reads from variant calling")
algorithm.add_argument('--no_delly','--no-delly',action="store_true",help="Don't run delly")
algorithm.add_argument('--no_lineage','--no-lineage',action="store_true",help=argparse.SUPPRESS)
algorithm.add_argument('--add_variant_annotations','--add-variant-annotations',action="store_true",help=argparse.SUPPRESS)
algorithm.add_argument('--threads','-t',default=1,help='Threads to use',type=int)
algorithm.add_argument('--ram',default=2,help='Maximum memory to use',type=int)
algorithm.add_argument('--implement_rules','--implement-rules',action="store_true",help='Use rules implemented in the resistance library (by default only a note will be made)')

plugins=parser_sub.add_argument_group("Plugin options")
for cls in tbp.ProfilePlugin.__subclasses__():
    if hasattr(cls,"__cli_params__"):
        for a in cls.__cli_params__:
            plugins.add_argument(*a['args'],**a['kwargs'])

other=parser_sub.add_argument_group("Other options")
other.add_argument('--logging',type=str.upper,default="INFO",choices=["DEBUG","INFO","WARNING","ERROR","CRITICAL"],help='Logging level')
other.add_argument('--debug',action='store_true',help=argparse.SUPPRESS)
other.add_argument('--delly_vcf','--delly-vcf',help=argparse.SUPPRESS)
other.add_argument('--supplementary_bam','--supplementary-bam',help=argparse.SUPPRESS)
other.add_argument('--low_dp_mask','--low-dp-mask',help=argparse.SUPPRESS)
other.add_argument('--save_low_dp_mask','--save-low-dp-mask',action='store_true',help=argparse.SUPPRESS)
other.add_argument('--no_clean','--no-clean', action='store_true',help=argparse.SUPPRESS)
other.add_argument('--temp',help="Temp firectory to process all files",type=str,default=".")
other.add_argument('--version', action='version', version="tb-profiler version %s" % tbp.__version__)

parser_sub.set_defaults(func=main_profile)



parser_sub = subparsers.add_parser('lineage', help='Profile only lineage', formatter_class=ArgumentDefaultsRichHelpFormatter)
parser_sub.add_argument('--bam','-a',required=True, help='BAM file. Make sure it has been generated using the H37Rv genome (GCA_000195955.2)')
parser_sub.add_argument('--prefix','-p',default="tbprofiler",help='Sample prefix')
parser_sub.add_argument('--barcode_snps','--barcode-snps',help='Dump barcoding mutations to a file')
parser_sub.add_argument('--caller',default='freebayes',choices=["bcftools","freebayes","gatk"],type=str,help="Variant caller")
parser_sub.add_argument('--kmer_counter','--kmer-counter',default='kmc',choices=["kmc","dsk"],type=str,help="Kmer counter")
parser_sub.add_argument('--platform','-m',choices=["illumina","nanopore","pacbio"],default="illumina",help='NGS Platform used to generate data')
parser_sub.add_argument('--db',default='tbdb',help='Mutation panel name')
parser_sub.add_argument('--spoligotype',action="store_true",help="Perform in-silico spoligotyping")
parser_sub.add_argument('--external_db','--external-db',type=str,help='Path to db files prefix (overrides "--db" parameter)')
parser_sub.add_argument('--text_template','--text-template',type=str,help='Jinja2 formatted template for output')
parser_sub.add_argument('--threads','-t',default=1,help='Threads to use',type=int)
parser_sub.add_argument('--dir','-d',default=".",help='Storage directory')
parser_sub.add_argument('--no_clean','--no-clean', action='store_true',help=argparse.SUPPRESS)
parser_sub.add_argument('--temp',help="Temp firectory to process all files",type=str,default=".")
parser_sub.add_argument('--version', action='version', version="tb-profiler version %s" % tbp.__version__)
parser_sub.add_argument('--logging',type=str.upper,default="INFO",choices=["DEBUG","INFO","WARNING","ERROR","CRITICAL"],help='Logging level')
parser_sub.add_argument('--debug',action='store_true',help=argparse.SUPPRESS)
parser_sub.set_defaults(func=main_lineage)

parser_sub = subparsers.add_parser('spoligotype', help='Profile spoligotype', formatter_class=ArgumentDefaultsRichHelpFormatter)
group = parser_sub.add_mutually_exclusive_group(required=True)
group.add_argument('--read1','-1',help='First read file')
parser_sub.add_argument('--read2','-2',help='Second read file')
group.add_argument('--bam','-a',help='BAM file. Make sure it has been generated using the H37Rv genome (GCA_000195955.2)')
group.add_argument('--fasta','-f',help='Fasta file')
parser_sub.add_argument('--prefix','-p',default="tbprofiler",help='Sample prefix')
parser_sub.add_argument('--txt',action="store_true",help="Add text output")
parser_sub.add_argument('--csv',action="store_true",help="Add CSV output")
parser_sub.add_argument('--docx',action="store_true",help="Add docx output. This requires docxtpl to be installed")
parser_sub.add_argument('--text_template',type=str,help='Jinja2 formatted template for output')
parser_sub.add_argument('--kmer_counter','--kmer-counter',default='kmc',choices=["kmc","dsk"],type=str,help="Kmer counter")
parser_sub.add_argument('--platform','-m',choices=["illumina","nanopore","pacbio"],default="illumina",help='NGS Platform used to generate data')
parser_sub.add_argument('--db',default='tbdb',help='Mutation panel name')
parser_sub.add_argument('--external_db','--external-db',type=str,help='Path to db files prefix (overrides "--db" parameter)')
parser_sub.add_argument('--dir','-d',default=".",help='Storage directory')
parser_sub.add_argument('--no_clean','--no-clean', action='store_true',help=argparse.SUPPRESS)
parser_sub.add_argument('--threads','-t',default=1,help='Threads to use',type=int)
parser_sub.add_argument('--ram',default=2,type=int_2_or_more,help='Maximum memory to use in Gb')
parser_sub.add_argument('--temp',help="Temp firectory to process all files",type=str,default=".")
parser_sub.add_argument('--version', action='version', version="tb-profiler version %s" % tbp.__version__)
parser_sub.add_argument('--logging',type=str.upper,default="INFO",choices=["DEBUG","INFO","WARNING","ERROR","CRITICAL"],help='Logging level')
parser_sub.add_argument('--debug',action='store_true',help=argparse.SUPPRESS)
parser_sub.set_defaults(func=main_spoligotype)

parser_sub = subparsers.add_parser('collate', help='Collate results form multiple samples together', formatter_class=ArgumentDefaultsRichHelpFormatter)
parser_sub.add_argument('--prefix','-p',default="tbprofiler",help='Sample prefix')
parser_sub.add_argument('--samples',help='File with samples (one per line)')
parser_sub.add_argument('--itol',action="store_true",help='Generate itol config files')
parser_sub.add_argument('--full',action="store_true",help='Output mutations in main result file')
parser_sub.add_argument('--all_variants','--all-variants',action="store_true",help='Output all variants in variant matrix')
parser_sub.add_argument('--mark_missing','--mark-missing',action="store_true",help='An asteriks will be use to mark predictions which are affected by missing data at a drug resistance position')
parser_sub.add_argument('--db',default='tbdb',help='Full path to mutation database json file to use')
parser_sub.add_argument('--format',default='txt',choices=['txt','csv'],help='Format of the output')
parser_sub.add_argument('--external_db','--external-db',type=str,help='Path to db files prefix (overrides "--db" parameter)')
parser_sub.add_argument('--dir','-d',nargs="+",default=["results"],help='Storage directory')
parser_sub.add_argument('--no_clean','--no-clean', action='store_true',help=argparse.SUPPRESS)
parser_sub.add_argument('--temp',help="Temp firectory to process all files",type=str,default=".")
parser_sub.add_argument('--version', action='version', version="tb-profiler version %s" % tbp.__version__)
parser_sub.add_argument('--logging',type=str.upper,default="INFO",choices=["DEBUG","INFO","WARNING","ERROR","CRITICAL"],help='Logging level')
parser_sub.add_argument('--debug',action='store_true',help=argparse.SUPPRESS)
parser_sub.set_defaults(func=main_collate)


parser_sub = subparsers.add_parser('reformat', help='Reformat json results into text or csv', formatter_class=ArgumentDefaultsRichHelpFormatter)
parser_sub.add_argument('json',default="tbprofiler",help='Sample prefix')
parser_sub.add_argument('--txt',action="store_true",help="Add text output")
parser_sub.add_argument('--csv',action="store_true",help="Add CSV output")
parser_sub.add_argument('--docx',action="store_true",help="Add docx output. This requires docxtpl to be installed")
parser_sub.add_argument('--docx_template','--docx-template',help="Supply custom template for --docx output")
parser_sub.add_argument('--docx_plugin','--docx-plugin',choices=list(docx_plugins),help="Use a plugin template for --docx output")
parser_sub.add_argument('--text_template','--text-template',type=str,help='Jinja2 formatted template for output')
parser_sub.add_argument('--db',default='tbdb',help='Mutation panel name')
parser_sub.add_argument('--external_db','--external-db',type=str,help='Path to db files prefix (overrides "--db" parameter)')
parser_sub.add_argument('--dir','-d',default=".",help='Storage directory')
parser_sub.add_argument('--no_clean','--no-clean', action='store_true',help=argparse.SUPPRESS)
parser_sub.add_argument('--suspect',action="store_true",help=argparse.SUPPRESS)
parser_sub.add_argument('--temp',help="Temp firectory to process all files",type=str,default=".")
parser_sub.add_argument('--version', action='version', version="tb-profiler version %s" % tbp.__version__)
parser_sub.add_argument('--logging',type=str.upper,default="INFO",choices=["DEBUG","INFO","WARNING","ERROR","CRITICAL"],help='Logging level')
parser_sub.add_argument('--debug',action='store_true',help=argparse.SUPPRESS)
parser_sub.set_defaults(func=main_reformat)

parser_sub = subparsers.add_parser('create_db', help='Generate the files required to run tb-profiler', formatter_class=ArgumentDefaultsRichHelpFormatter)
parser_sub.add_argument('--prefix','-p',type=str,help='The prefix for all output files',required=True)
parser_sub.add_argument('--csv','-c',nargs="+",default=["mutations.csv"],type=str,help='The input CSV file containing the mutations')
parser_sub.add_argument('--watchlist','-w',default="watchlist.csv",type=str,help='A csv file containing genes to profile but without any specific associated mutations')
parser_sub.add_argument('--spoligotypes',default="spoligotype_spacers.txt",type=str,help='A file containing a list of spoligotype spacers')
parser_sub.add_argument('--spoligotype_annotations','--spoligotype-annotations',default="spoligotype_list.csv")
parser_sub.add_argument('--barcode',default="barcode.bed",type=str,help='A bed file containing lineage barcode SNPs')
parser_sub.add_argument('--bedmask',default="mask.bed",type=str,help='A bed file containing a list of low-complexity regions')
parser_sub.add_argument('--rules',type=str,default="rules.txt",help='A file containing python rules')
parser_sub.add_argument('--amplicon_primers','--amplicon-primers',type=str,help='A file containing a list of amplicon primers')
parser_sub.add_argument('--match_ref','--match-ref',type=str,help='Match the chromosome name to the given fasta file')
parser_sub.add_argument('--custom',action="store_true",help='Tells the script this is a custom database, this is used to alter the generation of the version definition')
parser_sub.add_argument('--db_name',help='Overrides the name of the database in the version file')
parser_sub.add_argument('--db_commit','--db-commit',help='Overrides the commit string of the database in the version file')
parser_sub.add_argument('--db_author','--db-author',help='Overrides the author of the database in the version file')
parser_sub.add_argument('--db_date','--db-date',help='Overrides the date of the database in the version file')
parser_sub.add_argument('--include_original_mutation','--include-original-mutation',action="store_true", help='Include the original mutation (before reformatting) as part of the variant annotaion')
parser_sub.add_argument('--load',action="store_true", help='Automaticaly load database')
parser_sub.add_argument('--no_overwrite','--no-overwrite',action="store_true", help="Don't load if existing database with prefix exists")
parser_sub.add_argument('--dir','-d',default=".",help='Storage directory')
parser_sub.add_argument('--no_clean','--no-clean', action='store_true',help=argparse.SUPPRESS)
parser_sub.add_argument('--temp',help="Temp firectory to process all files",type=str,default=".")
parser_sub.add_argument('--version', action='version', version="tb-profiler version %s" % tbp.__version__)
parser_sub.add_argument('--logging',type=str.upper,default="INFO",choices=["DEBUG","INFO","WARNING","ERROR","CRITICAL"],help='Logging level')
parser_sub.add_argument('--debug',action='store_true',help=argparse.SUPPRESS)
parser_sub.set_defaults(func=main_create_db)

parser_sub = subparsers.add_parser('load_library', help='Load new library', formatter_class=ArgumentDefaultsRichHelpFormatter)
parser_sub.add_argument('prefix',type=str,help='Prefix to the library files')
parser_sub.add_argument('--dir','-d',default=".",help='Storage directory')
parser_sub.add_argument('--no_clean','--no-clean', action='store_true',help=argparse.SUPPRESS)
parser_sub.add_argument('--temp',help="Temp firectory to process all files",type=str,default=".")
parser_sub.add_argument('--version', action='version', version="tb-profiler version %s" % tbp.__version__)
parser_sub.add_argument('--logging',type=str.upper,default="INFO",choices=["DEBUG","INFO","WARNING","ERROR","CRITICAL"],help='Logging level')
parser_sub.add_argument('--debug',action='store_true',help=argparse.SUPPRESS)
parser_sub.set_defaults(func=main_load_library)

parser_sub = subparsers.add_parser('update_tbdb', help='Pull the latest tbdb library and load', formatter_class=ArgumentDefaultsRichHelpFormatter)
parser_sub.add_argument('--prefix','-p',help='Database name')
parser_sub.add_argument('--repo','-r',default="https://github.com/jodyphelan/tbdb.git",help='Repository to pull from')
parser_sub.add_argument('--branch','-b',default="tbdb",help='Branch to pull from')
parser_sub.add_argument('--commit','-c',help='Git commit hash to checkout (default: latest)')
parser_sub.add_argument('--match_ref','--match-ref',type=str,help='The prefix for all output files')
parser_sub.add_argument('--dir','-d',default=".",help='Storage directory')
parser_sub.add_argument('--no_clean','--no-clean', action='store_true',help=argparse.SUPPRESS)
parser_sub.add_argument('--temp',help="Temp firectory to process all files",type=str,default=".")
parser_sub.add_argument('--version', action='version', version="tb-profiler version %s" % tbp.__version__)
parser_sub.add_argument('--logging',type=str.upper,default="INFO",choices=["DEBUG","INFO","WARNING","ERROR","CRITICAL"],help='Logging level')
parser_sub.add_argument('--debug',action='store_true',help=argparse.SUPPRESS)
parser_sub.set_defaults(func=main_update_tbdb)

parser_sub = subparsers.add_parser('batch', help='Run tb-profiler for several samples', formatter_class=ArgumentDefaultsRichHelpFormatter)
parser_sub.add_argument('--csv',help='CSV with samples and files',required=True)
parser_sub.add_argument('--args',type=str, help='Arguments to use with tb-profiler')
parser_sub.add_argument('--jobs','-j',default=1,help='Threads to use',type=int)
parser_sub.add_argument('--threads_per_job','--threads-per-job','-t',default=1,help='Threads to use',type=int)
parser_sub.add_argument('--dir','-d',default=".",help='Storage directory')
parser_sub.add_argument('--no_clean','--no-clean', action='store_true',help=argparse.SUPPRESS)
parser_sub.add_argument('--temp',help="Temp firectory to process all files",type=str,default=".")
parser_sub.add_argument('--version', action='version', version="tb-profiler version %s" % tbp.__version__)
parser_sub.add_argument('--logging',type=str.upper,default="INFO",choices=["DEBUG","INFO","WARNING","ERROR","CRITICAL"],help='Logging level')
parser_sub.add_argument('--debug',action='store_true',help=argparse.SUPPRESS)
parser_sub.set_defaults(func=main_batch)

parser_sub = subparsers.add_parser('list_db', help='List loaded databases', formatter_class=ArgumentDefaultsRichHelpFormatter)
parser_sub.add_argument('--dir','-d',default=".",help='Storage directory')
parser_sub.add_argument('--no_clean','--no-clean', action='store_true',help=argparse.SUPPRESS)
parser_sub.add_argument('--temp',help="Temp firectory to process all files",type=str,default=".")
parser_sub.add_argument('--version', action='version', version="tb-profiler version %s" % tbp.__version__)
parser_sub.add_argument('--logging',type=str.upper,default="INFO",choices=["DEBUG","INFO","WARNING","ERROR","CRITICAL"],help='Logging level')
parser_sub.add_argument('--debug',action='store_true',help=argparse.SUPPRESS)
parser_sub.set_defaults(func=main_list_db)

parser_sub = subparsers.add_parser('version', help='Output program version and exit', formatter_class=ArgumentDefaultsRichHelpFormatter)
parser_sub.add_argument('--dir','-d',default=".",help='Storage directory')
parser_sub.add_argument('--no_clean','--no-clean', action='store_true',help=argparse.SUPPRESS)
parser_sub.add_argument('--temp',help="Temp firectory to process all files",type=str,default=".")
parser_sub.add_argument('--version', action='version', version="tb-profiler version %s" % tbp.__version__)
parser_sub.add_argument('--logging',type=str.upper,default="INFO",choices=["DEBUG","INFO","WARNING","ERROR","CRITICAL"],help='Logging level')
parser_sub.add_argument('--debug',action='store_true',help=argparse.SUPPRESS)
parser_sub.set_defaults(func=main_version)


args = parser.parse_args()

if args.debug:
    args.logging = "DEBUG"
logging.basicConfig(
    level=args.logging, format="%(message)s", datefmt="[%X]", handlers=[RichHandler()]
)

if hasattr(args, 'func'):
    args.software_name = __softwarename__
    args.version = tbp.__version__
    args.tmp_prefix = str(uuid4())
    args.files_prefix = os.path.abspath(f"{args.temp}/{args.tmp_prefix}")
    if hasattr(args,'dir'):
        if isinstance(args.dir,list):
            args.dir = [os.path.abspath(x) for x in args.dir]
        else:
            args.dir = os.path.abspath(args.dir)

    if hasattr(args, 'db'):
        if args.db=="tbdb" and not args.external_db and pp.nofile(sys.base_prefix+"/share/tbprofiler/tbdb.fasta"):
            logging.error("Can't find the tbdb file at %s. Please run 'tb-profiler update_tbdb' to load the default library or specify another using the '--external_db' flag" % sys.base_prefix)
            raise SystemExit
        if args.external_db:
            args.conf = pp.get_db(args.software_name,args.external_db)
        else:
            args.conf = pp.get_db(args.software_name,args.db)
        if args.conf is None:
            logging.error("Can't find the database %s. Please run 'tb-profiler create_db' to create the database or specify another using the '--external_db' flag" % args.db)
            raise SystemExit
        
    args.plugins = docx_plugins
    
    args.func(args)
else:
    parser.print_help(sys.stderr)