treesak 1.51.2__py3-none-any.whl

TreeSAK/Newick_tree_plotter.py

@@ -0,0 +1,79 @@
+import os
+from ete3 import Tree
+from ete3 import TreeStyle
+from ete3 import NodeStyle
+from ete3 import TextFace
+
+
+def plot_tree(tree, tree_title, tree_output):
+    # set tree parameters
+    ts = TreeStyle()
+    ts.mode = "r"  # tree model: 'r' for rectangular, 'c' for circular
+    ts.show_leaf_name = 0
+    # set tree title text parameters
+    ts.title.add_face(TextFace(tree_title,
+                               fsize = 8,
+                               fgcolor = 'black',
+                               ftype = 'Arial',
+                               tight_text = False),
+                      column = 0)  # tree title text setting
+    # set layout parameters
+    ts.rotation = 0  # from 0 to 360
+    ts.show_scale = False
+    ts.margin_top = 10  # top tree image margin
+    ts.margin_bottom = 10  # bottom tree image margin
+    ts.margin_left = 10  # left tree image margin
+    ts.margin_right = 10  # right tree image margin
+    ts.show_border = False  # set tree image border
+    ts.branch_vertical_margin = 3  # 3 pixels between adjancent branches
+
+    # set tree node style
+    for each_node in tree.traverse():
+        # leaf node parameters
+        if each_node.is_leaf():
+            ns = NodeStyle()
+            ns["shape"] = "circle"  # dot shape: circle, square or sphere
+            ns["size"] = 0  # dot size
+            ns['hz_line_width'] = 0.5  # branch line width
+            ns['vt_line_width'] = 0.5  # branch line width
+            ns['hz_line_type'] = 0  # branch line type: 0 for solid, 1 for dashed, 2 for dotted
+            ns['vt_line_type'] = 0  # branch line type
+            ns["fgcolor"] = "blue"  # the dot setting
+            each_node.add_face(TextFace(each_node.name,
+                                        fsize = 5,
+                                        fgcolor = 'black',
+                                        tight_text = False,
+                                        bold = False),
+                               column = 0,
+                               position = 'branch-right')  # leaf node the node name text setting
+
+            each_node.set_style(ns)
+
+        # non-leaf node parameters
+        else:
+            nlns = NodeStyle()
+            nlns["size"] = 0  # dot size
+            #nlns["rotation"] = 45
+            each_node.add_face(TextFace(each_node.name,
+
+                                        fsize = 3,
+                                        fgcolor = 'black',
+                                        tight_text = False,
+                                        bold = False),
+                               column = 5,
+                               position = 'branch-top')  # non-leaf node name text setting)
+
+            each_node.set_style(nlns)
+
+    tree.render(tree_output, w=900, units="px", tree_style=ts)  # set figures size
+
+
+#os.chdir('/Users/songweizhi/Desktop')
+
+tree_2 = '(CF_Refined_71:0.21847,CF_Refined_170:0.41504,(((CF_Refined_7:0.63495,CF_Refined_96:0.68718)0.984:0.33915,CF_Refined_82:0.16074)0.980:0.12012,((CF_Refined_25:0.20437,CF_Refined_95:1.40476)0.367:0.60450,(((CF_Refined_86:0.37933,(CF_Refined_74:0.61406,CF_Refined_43:0.10850)1.000:0.34468)0.999:0.19175,((CF_Refined_57:0.19003,(CF_Refined_99:0.18534,CF_Refined_160:0.33153)0.861:0.04660)1.000:0.18553,(CF_Refined_78:0.64747,(CF_Refined_129:0.26317,(CF_Refined_64:0.25293,CF_Refined_100:0.10449)0.993:0.14949)0.577:0.13006)1.000:0.19231)0.998:0.16057)0.961:0.08413,((CF_Refined_155:0.18262,CF_Refined_180:0.02711)1.000:0.42318,(CF_Refined_162:0.64092,CF_Refined_131:0.36385)1.000:0.48656)0.992:0.23471)0.853:0.05581)0.955:0.08666)1.000:0.14706);'
+
+tree = Tree(tree_2, format=1)
+
+
+plot_tree(tree, 'Species_Tree', '/Users/songweizhi/Desktop/Species_Tree.png')
+

TreeSAK/OMA.py

@@ -0,0 +1,170 @@
+import os
+import glob
+import argparse
+
+
+OMA_usage = '''
+======================= OMA example commands =======================
+
+TreeSAK OMA -i faa_files -x faa -og og_gnm.txt -o OMA_wd -f -t 32
+
+====================================================================
+'''
+
+def sep_path_basename_ext(file_in):
+
+    f_path, file_name = os.path.split(file_in)
+    if f_path == '':
+        f_path = '.'
+    f_base, f_ext = os.path.splitext(file_name)
+    return f_path, f_base, f_ext
+
+
+def get_default_para_dict():
+
+    default_para_str = '''
+    OutputFolder := 'Output';
+    ReuseCachedResults := true;
+    AlignBatchSize := 1e6;
+    MinScore := 181;
+    LengthTol := 0.61;
+    StablePairTol := 1.81;
+    InparalogTol := 3.00;
+    ParalogTol := -2.5*StablePairTol;
+    VerifiedPairTol := 1.53;
+    MinSeqLen := 50;
+    UseOnlyOneSplicingVariant := true;
+    UseExperimentalHomologousClusters := false;
+    QuasiCliquesCutoff := 1.0:
+    StableIdsForGroups := false;
+    GuessIdType := false;
+    DoHierarchicalGroups := 'bottom-up';
+    SpeciesTree := 'estimate';
+    MinEdgeCompletenessFraction := 0.65;
+    ReachabilityCutoff := 0.65;
+    MaxTimePerLevel := 1200;  # 20min
+    DoGroupFunctionPrediction := true;
+    GroupFunctionCutoff := 0.5;
+    CladeDefinition := 'default';
+    UseEsprit := false;
+    DistConfLevel := 2;
+    MinProbContig := 0.4;
+    MaxContigOverlap := 5;
+    MinSeqLenContig := 20;
+    MinBestScore := 250;
+    '''
+
+    default_para_dict = dict()
+    for each_line in default_para_str.split('    '):
+        para_line = each_line.replace(' ', '').replace('\n', '').split(';')[0]
+        if para_line != '':
+            para_line_split = para_line.split(':=')
+            default_para_dict[para_line_split[0]] = para_line_split[1]
+
+    return default_para_dict
+
+
+def OMA(args):
+
+    gnm_dir         = args['i']
+    file_ext        = args['x']
+    seq_type        = args['st']
+    og_gnm_txt      = args['og']
+    op_dir          = args['o']
+    force_overwrite = args['f']
+    num_threads     = args['t']
+
+    # define file name
+    pwd_gnm_rename_txt = '%s/rename.txt'     % op_dir
+    pwd_parameter_file = '%s/parameters.drw' % op_dir
+    oma_input_dir      = '%s/DB'             % op_dir
+
+    # create dir
+    if os.path.isdir(op_dir) is True:
+        if force_overwrite is True:
+            os.system('rm -r %s' % op_dir)
+        else:
+            print('output folder detected, program exited!')
+            exit()
+    os.system('mkdir %s' % op_dir)
+    os.system('mkdir %s' % oma_input_dir)
+
+    # check genome files
+    gnm_file_re   = '%s/*.%s' % (gnm_dir, file_ext)
+    gnm_file_list = glob.glob(gnm_file_re)
+    if len(gnm_file_list) == 0:
+        print('No genome detected, program exited!')
+        exit()
+
+    # check og_gnm_txt
+    if os.path.isfile(og_gnm_txt) is False:
+        print('Out group genome id file not detected, program exited!')
+        exit()
+
+    # copy genome files into DB folder
+    gnm_id_rename_dict = dict()
+    rename_list = []
+    for each_gnm in gnm_file_list:
+        gnm_path, gnm_base, gnm_ext = sep_path_basename_ext(each_gnm)
+        gnm_base_renamed = gnm_base.replace('.', '_')
+        pwd_gnm_db = '%s/%s.fa' % (oma_input_dir, gnm_base_renamed)
+        if gnm_base != gnm_base_renamed:
+            rename_list.append('%s\t%s' % (gnm_base, gnm_base_renamed))
+        gnm_id_rename_dict[gnm_base] = gnm_base_renamed
+        os.system('cp %s %s' % (each_gnm, pwd_gnm_db))
+
+    # write out rename file
+    if len(rename_list) > 0:
+        pwd_gnm_rename_txt_handle = open(pwd_gnm_rename_txt, 'w')
+        for each_e in sorted(rename_list):
+            pwd_gnm_rename_txt_handle.write(each_e + '\n')
+        pwd_gnm_rename_txt_handle.close()
+    else:
+        print('Format of file names passed checking')
+
+    # get default_para_dict
+    default_para_dict = get_default_para_dict()
+
+    # read in og_gnm_txt
+    renamed_og_gnm_list = []
+    for each_og_gnm in open(og_gnm_txt):
+        og_gnm_renamed = gnm_id_rename_dict[each_og_gnm.strip()]
+        renamed_og_gnm_list.append(og_gnm_renamed)
+
+    # write out parameter file
+    with open(pwd_parameter_file, 'w') as pwd_parameter_file_handle:
+
+        # write InputDataType line
+        if seq_type in ['AA', 'aa', 'Aa']:
+            pwd_parameter_file_handle.write("InputDataType := 'AA';\n")
+        if seq_type in ['DNA', 'dna', 'Dna']:
+            pwd_parameter_file_handle.write("InputDataType := 'DNA';\n")
+
+        # write OutgroupSpecies line
+        OutgroupSpecies_value_str = "['%s']" % "', '".join(renamed_og_gnm_list)
+        pwd_parameter_file_handle.write("OutgroupSpecies := %s;\n" % OutgroupSpecies_value_str)
+
+        # write out the rest lines
+        for each_para in default_para_dict:
+            para_value = default_para_dict[each_para]
+            pwd_parameter_file_handle.write("%s := %s;\n" % (each_para, para_value))
+
+    # final report
+    print('You can run OMA with:')
+    print('cd %s' % op_dir)
+    print('oma -n %s' % num_threads)
+    print('# You may want to customize parameters specified in %s ' % pwd_parameter_file)
+
+
+if __name__ == '__main__':
+
+    OMA_parser = argparse.ArgumentParser()
+    OMA_parser.add_argument('-i',   required=True,                       help='genome folder')
+    OMA_parser.add_argument('-x',   required=True,                       help='genome file extension')
+    OMA_parser.add_argument('-st',  required=False, default='AA',        help='sequence type, AA or DNA, default: AA')
+    OMA_parser.add_argument('-og',  required=True,                       help='outgroup genomes, without file extension')
+    OMA_parser.add_argument('-o',   required=True,  default=None,        help='output dir, i.e., OMA working directory')
+    OMA_parser.add_argument('-f',   required=False, action="store_true", help='force overwrite')
+    OMA_parser.add_argument('-t',   required=False, type=int, default=6, help='number of threads for running OMA, default: 6')
+    args = vars(OMA_parser.parse_args())
+    OMA(args)

TreeSAK/OMA2.py

@@ -0,0 +1,212 @@
+import os
+import glob
+import argparse
+from Bio import SeqIO
+
+
+OMA2_usage = '''
+============================== OMA2 example commands ==============================
+
+TreeSAK OMA2 -i OrthologousGroups.txt -s OrthologousGroupsFasta -o op_dir -f -n 3
+TreeSAK OMA2 -i OrthologousGroups.txt -s OrthologousGroupsFasta -o op_dir -f -c 85
+
+===================================================================================
+'''
+
+
+def sep_path_basename_ext(file_in):
+
+    f_path, f_name = os.path.split(file_in)
+    if f_path == '':
+        f_path = '.'
+    f_base, f_ext = os.path.splitext(f_name)
+
+    return f_name, f_path, f_base, f_ext[1:]
+
+
+def get_gnm_og_cov(og_dir, og_ext, og_cov_txt):
+
+    og_file_re   = '%s/*.%s' % (og_dir, og_ext)
+    og_file_list = glob.glob(og_file_re)
+
+    gnm_to_og_dict = dict()
+    for og_file in og_file_list:
+        _, _, og_id, _ = sep_path_basename_ext(og_file)
+        for each_seq in SeqIO.parse(og_file, 'fasta'):
+            seq_id = each_seq.id
+            gnm_id = '_'.join(seq_id.split('_')[:-1])
+            if gnm_id not in gnm_to_og_dict:
+                gnm_to_og_dict[gnm_id] = set()
+            gnm_to_og_dict[gnm_id].add(og_id)
+
+    og_cov_txt_handle = open(og_cov_txt, 'w')
+    for each_gnm in sorted(list(gnm_to_og_dict.keys())):
+        gnm_og_set = gnm_to_og_dict[each_gnm]
+        og_cov = len(gnm_og_set)*100/len(og_file_list)
+        og_cov = float("{0:.2f}".format(og_cov))
+        og_cov_txt_handle.write('%s\t%s\n' % (each_gnm, og_cov))
+    og_cov_txt_handle.close()
+
+
+def get_ortho_to_gene_dict(ortho_groups_txt, og_program):
+
+    ortho_to_gene_dict = dict()
+    for each_og in open(ortho_groups_txt):
+        if not each_og.startswith('#'):
+            og_id = ''
+            gene_list = []
+            if og_program == 'orthofinder':
+                each_og_split = each_og.strip().split(' ')
+                og_id = each_og_split[0][:-1]
+                gene_list = each_og_split[1:]
+            elif og_program == 'oma':
+                each_og_split = each_og.strip().split('\t')
+                og_id = each_og_split[0]
+                group_member_list = each_og_split[1:]
+                for each_protein in group_member_list:
+                    protein_id = each_protein.split(' ')[0].split(':')[1]
+                    gene_list.append(protein_id)
+            ortho_to_gene_dict[og_id] = gene_list
+
+    return ortho_to_gene_dict
+
+
+def select_seq(seq_in, seq_id_list, seq_out):
+    output_file_handle = open(seq_out, 'w')
+    for seq_record in SeqIO.parse(seq_in, 'fasta'):
+        if seq_record.id in seq_id_list:
+            output_file_handle.write('>%s\n' % seq_record.id)
+            output_file_handle.write('%s\n' % seq_record.seq)
+    output_file_handle.close()
+
+
+def OMA2(args):
+
+    og_txt              = args['i']
+    og_seq_dir          = args['s']
+    gnm_txt             = args['g']
+    op_dir              = args['o']
+    force_overwrite     = args['f']
+    min_gene_num        = args['n']
+    min_gene_cov        = args['c']
+
+    if (min_gene_num is None) and (min_gene_cov is None):
+        print('Please specify either -c or -n, program exited!')
+        exit()
+    elif (min_gene_num is not None) and (min_gene_cov is not None):
+        print('-c and -n are not compatible, program exited!')
+        exit()
+
+    og_txt_out = ''
+    gnm_og_num_txt = ''
+    filtered_seq_dir = ''
+    if min_gene_num is not None:
+        og_txt_out       = '%s/OrthologousGroups_num%s.txt'             % (op_dir, min_gene_num)
+        gnm_og_num_txt   = '%s/OrthologousGroups_num%s_per_genome.txt'  % (op_dir, min_gene_num)
+        filtered_seq_dir = '%s/OrthologousGroupsFasta_num%s'            % (op_dir, min_gene_num)
+    if min_gene_cov is not None:
+        og_txt_out       = '%s/OrthologousGroups_cov%s.txt'             % (op_dir, min_gene_cov)
+        gnm_og_num_txt   = '%s/OrthologousGroups_cov%s_per_genome.txt'  % (op_dir, min_gene_cov)
+        filtered_seq_dir = '%s/OrthologousGroupsFasta_cov%s'            % (op_dir, min_gene_cov)
+
+    # check genome files
+    interested_gnm_set = set()
+    if gnm_txt is not None:
+        if os.path.isfile(gnm_txt) is True:
+            for each_gnm in open(gnm_txt):
+                gnm_id = each_gnm.strip().split()[0]
+                interested_gnm_set.add(gnm_id)
+        else:
+            print('%s not found, program exited!' % gnm_txt)
+            exit()
+
+    # create dir
+    if os.path.isdir(op_dir) is True:
+        if force_overwrite is True:
+            os.system('rm -r %s' % op_dir)
+        else:
+            print('output folder detected, program exited!')
+            exit()
+    os.system('mkdir %s' % op_dir)
+    os.system('mkdir %s' % filtered_seq_dir)
+
+    # get overall genome set
+    overall_gnm_set = set()
+    for each_line in open(og_txt):
+        if not each_line.startswith('#'):
+            each_line_split = each_line.strip().split('\t')
+            gene_list = each_line_split[1:]
+            for each_gene in gene_list:
+                gene_gnm = each_gene.split(':')[0]
+                overall_gnm_set.add(gene_gnm)
+
+    qualified_og_set = set()
+    id_to_name_dict = dict()
+    gene_to_extract_dict = dict()
+    og_txt_out_handle = open(og_txt_out, 'w')
+    for each_line in open(og_txt):
+        if not each_line.startswith('#'):
+            each_line_split = each_line.strip().split('\t')
+            og_id = each_line_split[0]
+            filename = 'OG%s' % int(og_id[3:])
+            id_to_name_dict[og_id] = filename
+            gene_list = each_line_split[1:]
+            filtered_gene_set = set()
+            for each_gene in gene_list:
+                gene_gnm = each_gene.split(':')[0]
+                gene_id = each_gene.split(':')[1].split(' ')[0]
+                if len(interested_gnm_set) == 0:
+                    filtered_gene_set.add(gene_id)
+                else:
+                    if gene_gnm in interested_gnm_set:
+                        filtered_gene_set.add(gene_id)
+
+            qualified_og = False
+            if min_gene_num is not None:
+                if len(filtered_gene_set) >= float(min_gene_num):
+                    qualified_og = True
+            if min_gene_cov is not None:
+                if len(interested_gnm_set) == 0:
+                    gnm_cov = len(filtered_gene_set)*100/len(overall_gnm_set)
+                else:
+                    gnm_cov = len(filtered_gene_set)*100/len(interested_gnm_set)
+
+                if gnm_cov >= float(min_gene_cov):
+                    qualified_og = True
+
+            if qualified_og is True:
+                qualified_og_set.add(og_id)
+                og_txt_out_handle.write('%s\t%s\n' % (filename, ','.join(sorted(list(filtered_gene_set)))))
+                gene_to_extract_dict[og_id] = filtered_gene_set
+    og_txt_out_handle.close()
+
+    for each_og in gene_to_extract_dict:
+        seq_file_name = id_to_name_dict[each_og]
+        source_file   = '%s/%s.fa' % (og_seq_dir, seq_file_name)
+        filtered_file = '%s/%s.fa' % (filtered_seq_dir, seq_file_name)
+        select_seq(source_file, gene_to_extract_dict[each_og], filtered_file)
+
+    # get_gnm_og_cov
+    get_gnm_og_cov(filtered_seq_dir, 'fa', gnm_og_num_txt)
+
+    # report
+    if min_gene_num is not None:
+        print('The number of OG with genes >= %s is %s' % (min_gene_num, len(qualified_og_set)))
+    if min_gene_cov is not None:
+        print('The number of OG with coverage >= %s is %s' % (min_gene_cov, len(qualified_og_set)))
+
+    print('Done!')
+
+
+if __name__ == '__main__':
+
+    OMA2_parser = argparse.ArgumentParser()
+    OMA2_parser.add_argument('-i',   required=True,                         help='OrthologousGroups.txt')
+    OMA2_parser.add_argument('-s',   required=True,                         help='sequence dir, OrthologousGroupsFasta')
+    OMA2_parser.add_argument('-g',   required=False, default=None,          help='interested genomes')
+    OMA2_parser.add_argument('-o',   required=True,  default=None,          help='output directory')
+    OMA2_parser.add_argument('-n',   required=False, default=None,          help='minimal number of gene in a OG, not compatible with -c')
+    OMA2_parser.add_argument('-c',   required=False, default=None,          help='minimal genome coverage cutoff, not compatible with -n')
+    OMA2_parser.add_argument('-f',   required=False, action="store_true",   help='force overwrite')
+    args = vars(OMA2_parser.parse_args())
+    OMA2(args)

TreeSAK/OneLineAln.py

@@ -0,0 +1,50 @@
+import argparse
+from Bio import SeqIO
+
+
+OneLineAln_usage = '''
+========================= OneLineAln example commands =========================
+
+BioSAK OneLineAln -in MarkerGenes.aln -out MarkerGenes_OneLine.aln
+BioSAK OneLineAln -in MarkerGenes.aln -out MarkerGenes_OneLine.aln -upper
+
+===============================================================================
+'''
+
+def OneLineAln(args):
+
+    aln_in_fasta     = args['in']
+    aln_out_one_line = args['out']
+    to_uppercase     = args['upper']
+
+    # get longest_seq_id
+    longest_seq_id = 0
+    for seq in SeqIO.parse(aln_in_fasta, 'fasta'):
+        if len(seq.id) > longest_seq_id:
+            longest_seq_id = len(seq.id)
+
+    # write out in new format
+    aln_in_one_line_handle = open(aln_out_one_line, 'w')
+    for seq in SeqIO.parse(aln_in_fasta, 'fasta'):
+        seq_id_polished = seq.id + (longest_seq_id - len(seq.id))*' '
+        if to_uppercase is True:
+            aln_in_one_line_handle.write('%s\t%s\n' % (seq_id_polished, str(seq.seq).upper()))
+        else:
+            aln_in_one_line_handle.write('%s\t%s\n' % (seq_id_polished, str(seq.seq)))
+
+    aln_in_one_line_handle.close()
+
+
+if __name__ == '__main__':
+
+    OneLineAln_parser = argparse.ArgumentParser()
+
+    # arguments for rename_seq_parser
+    OneLineAln_parser.add_argument('-in',     required=True,                        help='input MSA in fasta format')
+    OneLineAln_parser.add_argument('-out',    required=False, default=None,         help='output file')
+    OneLineAln_parser.add_argument('-upper',  required=False, action='store_true',  help='turn to uppercase')
+
+    args = vars(OneLineAln_parser.parse_args())
+
+    OneLineAln(args)
+

TreeSAK/PB.py

@@ -0,0 +1,155 @@
+import os
+import argparse
+from Bio import AlignIO
+
+
+PB_usage = '''
+========================================== PB example commands ==========================================
+
+# Dependency: mpirun, pb_mpi and readpb_mpi (from PhyloBayes-MPI)
+
+export OMPI_MCA_btl=^openib
+TreeSAK PB -i in.phylip -p best20pb -o best20pb -t 52 
+TreeSAK PB -i in.phylip -p best20pb -o best20pb -t 52 -n 1
+TreeSAK PB -i in.phylip -p worst20pb -o worst20pb -t 52 
+
+# Notes:
+1. This is a wrapper for: mpirun -np 12 pb_mpi -d in.phylip -cat -gtr -x 10 -1 -dgam 4 -s chain_name
+2. Input MSA need to be in phylip format.
+3. To stop a chain, just open the <chain_name>.run file and replace the 1 by a 0 (echo 0 > <chain_name>.run).
+4. Be careful not to restart an already running chain.
+5. You can stop a chain and restart it under a different degree of parallelization.
+6. Generally, PhyloBayes provides good results for a total number of points of 10000-30000.
+7. Results can be assessed with bpcomp and tracecomp
+
+*  Settings used by Nina Dombrowski: -cat -gtr -x 10 -1 -dgam 4
+   For each marker protein family, four parallel chains were run until convergence was reached, unless stated 
+   otherwise (maxdiff < 0.3; settings: bpcomp -x 25_burnin chain1 chain2 chain3 chain4). Additionally, we 
+   checked for the minimum effective size using tracecomp (minimum effective size > 50; settings: -x 25_burnin 
+   chain1 chain2 chain3 chain4).
+   
+*  Settings used by Fan Lu:
+   Four chains were run for each consensus tree, and for each chain over 15,000 cycles (5,000 burn-in) 
+   were conducted, until a maxdiff value lower than 0.3 was reached. Otherwise, non-converged chains were 
+   continually run to over 20,000 cycles. Posterior predictive tests were conducted using PhyloBayes MPI 
+   with the ‘readpb_mpi -x 5000 50 -allppred’ command.
+
+=========================================================================================================
+'''
+
+
+def sep_path_basename_ext(file_in):
+
+    f_path, f_name = os.path.split(file_in)
+    if f_path == '':
+        f_path = '.'
+    f_base, f_ext = os.path.splitext(f_name)
+
+    return f_name, f_path, f_base, f_ext[1:]
+
+
+def fa2phy(fasta_in, phy_out):
+
+    alignment = AlignIO.read(fasta_in, 'fasta')
+
+    max_seq_id_len = 0
+    for each_seq in alignment:
+        seq_id_len = len(each_seq.id)
+        if seq_id_len > max_seq_id_len:
+            max_seq_id_len = seq_id_len
+
+    with open(phy_out, 'w') as msa_out_handle:
+        msa_out_handle.write('%s %s\n' % (len(alignment), alignment.get_alignment_length()))
+        for each_seq in alignment:
+            seq_id = each_seq.id
+            seq_id_with_space = '%s%s' % (seq_id, ' ' * (max_seq_id_len + 2 - len(seq_id)))
+            msa_out_handle.write('%s%s\n' % (seq_id_with_space, str(each_seq.seq)))
+
+
+def PB(args):
+
+    msa_in          = args['i']
+    op_dir          = args['o']
+    op_prefix       = args['p']
+    fa_to_plp       = args['fa2plp']
+    num_of_threads  = args['t']
+    num_of_chains   = args['n']
+    force_overwrite = args['f']
+
+    ####################################################################################################################
+
+    msa_in_name, msa_in_path, msa_in_base, msa_in_ext = sep_path_basename_ext(msa_in)
+
+    settings_dombrowski = '-cat -gtr -x 10 -1 -dgam 4'
+    setting_to_use  = settings_dombrowski
+    msa_in_plp      = '%s/%s.phylip'        % (op_dir, msa_in_base)
+    cmd_txt         = '%s/%s_cmds.txt'      % (op_dir, msa_in_base)
+
+    ####################################################################################################################
+
+    # create output dir
+    if os.path.isdir(op_dir) is True:
+        if force_overwrite is True:
+            os.system('rm -r %s' % op_dir)
+        else:
+            print('output folder already exist, program exited!')
+            exit()
+    os.system('mkdir %s' % op_dir)
+
+    # fa_to_phylip
+    msa_to_use = msa_in
+    if fa_to_plp is True:
+        fa2phy(msa_in, msa_in_plp)
+        msa_to_use = msa_in_plp
+
+    chain_name_list = []
+    pb_mpi_cmd_list = []
+    if num_of_chains == 1:
+        pb_mpi_cmd = 'export OMPI_MCA_btl=^openib; mpirun -np %s pb_mpi -d %s %s -s %s/%s' % (num_of_threads, msa_to_use, setting_to_use, op_dir, op_prefix)
+        chain_name_list.append('%s/%s' % (op_dir, op_prefix))
+        pb_mpi_cmd_list.append(pb_mpi_cmd)
+    else:
+        for chain_index in range(1, (num_of_chains + 1)):
+            current_wd = '%s/%s_chain%s' % (op_dir, op_prefix, chain_index)
+            os.mkdir(current_wd)
+            pb_mpi_cmd = 'export OMPI_MCA_btl=^openib; mpirun -np %s pb_mpi -d %s %s -s %s/%s_chain%s' % (num_of_threads, msa_to_use, setting_to_use, current_wd, op_prefix, chain_index)
+            chain_name_list.append('%s/%s_chain%s' % (current_wd, op_prefix, chain_index))
+            pb_mpi_cmd_list.append(pb_mpi_cmd)
+
+    # write out commands
+    cmd_txt_handle = open(cmd_txt, 'w')
+    cmd_txt_handle.write('# To run pb_mpi\n')
+    for cmd in pb_mpi_cmd_list:
+        cmd_txt_handle.write(cmd + '\n')
+
+    cmd_txt_handle.write('\n# To restart a terminated run (e.g., due to walltime limitation)\n')
+    for each_chain in chain_name_list:
+        cmd_txt_handle.write('export OMPI_MCA_btl=^openib; mpirun -np %s pb_mpi %s\n' % (num_of_threads, each_chain))
+    cmd_txt_handle.close()
+
+    # assess the results
+    if num_of_chains > 1:
+        readpb_cmd = 'export OMPI_MCA_btl=^openib; bpcomp -x 1000 10 %s' % (' '.join(chain_name_list))
+        bpcomp_cmd = 'export OMPI_MCA_btl=^openib; tracecomp -x 1000 %s' % (' '.join(chain_name_list))
+        cmd_txt_handle = open(cmd_txt, 'a')
+        cmd_txt_handle.write('\n# You may want to use the following commands to assess the results:\n')
+        cmd_txt_handle.write(readpb_cmd + '\n')
+        cmd_txt_handle.write(bpcomp_cmd + '\n')
+        cmd_txt_handle.close()
+
+    print('Commands exported to %s' % cmd_txt)
+    print('Done!')
+
+
+if __name__ == '__main__':
+
+    PB_parser = argparse.ArgumentParser()
+    PB_parser.add_argument('-i',       required=True,                          help='input MSA file')
+    PB_parser.add_argument('-o',       required=True,                          help='output directory')
+    PB_parser.add_argument('-p',       required=True,                          help='output prefix')
+    PB_parser.add_argument('-fa2plp',  required=False, action="store_true",    help='convert MSA format from fasta to phylip')
+    PB_parser.add_argument('-n',       required=False, type=int, default=4,    help='number of chains, default: 4')
+    PB_parser.add_argument('-t',       required=False, type=int, default=48,   help='num of cores per mpirun, default: 48')
+    PB_parser.add_argument('-f',       required=False, action="store_true",    help='force overwrite')
+    args = vars(PB_parser.parse_args())
+    PB(args)