treesak 1.53.3__py3-none-any.whl

TreeSAK/SplitScore.py

@@ -0,0 +1,21 @@
+
+SplitScore_usage = '''
+============================================= SplitScore example commands =============================================
+
+# SplitScore modules
+TreeSAK SplitScore1     ->  Step 1: Infer gene tree
+TreeSAK SplitScore1OMA  ->  Step 1: Infer gene tree (based on OMA outputs)
+TreeSAK SplitScore2     ->  Step 2: Calculate split score
+
+# SplitScore1
+TreeSAK SplitScore1 -i OrthologousGroups.txt -s OrthologousGroupsFasta -o step1_op_dir -t 6 -f
+TreeSAK SplitScore1 -i OrthologousGroups.txt -s OrthologousGroupsFasta -o step1_op_dir -t 6 -f -u interested_gnm.txt
+
+# SplitScore2
+# Please ensure that all the commands produced in step one have been executed before proceeding to step two.
+TreeSAK SplitScore2 -i step1_op_dir -g gnm_cluster.tsv -k gnm_taxon.txt -f -t 10 -o step_2_op_dir
+
+# As described in the Undinarchaeota paper (Nina Dombrowski 2020, NC)
+
+=======================================================================================================================
+'''

TreeSAK/SplitScore1.py

@@ -0,0 +1,177 @@
+from __future__ import print_function
+import os
+import glob
+import argparse
+from Bio import SeqIO
+
+
+SplitScore1_usage = '''
+======================== SplitScore1 example commands ========================
+
+TreeSAK SplitScore1 -i marker_seq -x fa -o SplitScore1_op_dir -jst 9 -f
+
+# As described in the Undinarchaeota paper (Nina Dombrowski 2020, NC)
+
+==============================================================================
+'''
+
+
+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 SplitScore1(args):
+
+    oma_op_fasta        = args['i']
+    fasta_file_ext      = args['x']
+    interested_gnm_txt  = args['u']
+    iqtree_model        = args['m']
+    cov_cutoff          = args['c']
+    force_overwrite     = args['f']
+    num_of_js_threads   = args['jst']
+    op_dir              = args['o']
+    seq_named_by_gnm    = args['seq_named_by_gnm']
+    bmge_trim_model             = 'BLOSUM30'
+    bmge_entropy_score_cutoff   = '0.55'
+
+    ################################################################################
+
+    interested_gnm_set = set()
+    if interested_gnm_txt is not None:
+        if os.path.isfile(interested_gnm_txt):
+            for each_gnm in open(interested_gnm_txt):
+                interested_gnm_set.add(each_gnm.strip())
+        else:
+            print('%s not found, program exited' % interested_gnm_txt)
+            exit()
+
+    ################################################################################
+
+    # specify path to BMGE.jar
+    current_file_path = '/'.join(os.path.realpath(__file__).split('/')[:-1])
+    pwd_bmge_jar      = '%s/BMGE.jar' % current_file_path
+
+    fa_file_re   = '%s/*.%s' % (oma_op_fasta, fasta_file_ext)
+    fa_file_list = glob.glob(fa_file_re)
+    if len(fa_file_list) == 0:
+        print('No file found in %s, program exited!' % oma_op_fasta)
+        exit()
+
+    og_to_gene_dict = dict()
+    for each_fa in fa_file_list:
+        _, f_base, _ = sep_path_basename_ext(each_fa)
+        seq_id_set = set()
+        for each_seq in SeqIO.parse(each_fa, 'fasta'):
+            seq_id_set.add(each_seq.id)
+        og_to_gene_dict[f_base] = seq_id_set
+
+    ################################################################################
+
+    gnm_to_process = set()
+    for each_og in og_to_gene_dict:
+        gene_set = og_to_gene_dict[each_og]
+        gnm_set = set()
+        for each_gene in gene_set:
+            gnm_id = '_'.join(each_gene.split('_')[:-1])
+            if seq_named_by_gnm is True:
+                gnm_id = each_gene
+            gnm_set.add(gnm_id)
+            if interested_gnm_txt is None:
+                gnm_to_process.add(gnm_id)
+            else:
+                if gnm_id in interested_gnm_set:
+                    gnm_to_process.add(gnm_id)
+
+        if len(gene_set) != len(gnm_set):
+            print('Program exited!')
+            exit()
+
+    ################################################################################
+
+    # define file name
+    qualified_og_dir        = '%s/qualified_OGs'                % op_dir
+    cmds_in_one_line_txt    = '%s/cmds_mafft_bmge_iqtree.txt'   % op_dir
+    ignored_marker_txt      = '%s/ignored_markers.txt'          % op_dir
+
+    # create output folder
+    if os.path.isdir(op_dir) is True:
+        if force_overwrite is True:
+            os.system('rm -r %s' % op_dir)
+        else:
+            print('%s exist, program exited!' % op_dir)
+            exit()
+    os.mkdir(op_dir)
+    os.mkdir(qualified_og_dir)
+
+    ################################################################################
+
+    cmds_in_one_line_txt_handle = open(cmds_in_one_line_txt, 'w')
+    ignored_og_dict = dict()
+    for each_og in sorted(list(og_to_gene_dict.keys())):
+        seq_file_in          = '%s/%s.%s'       % (oma_op_fasta, each_og, fasta_file_ext)
+        file_out_seq         = '%s/%s.%s'       % (qualified_og_dir, each_og, fasta_file_ext)
+        file_out_aln         = '%s.aln'         % each_og
+        file_out_aln_trimmed = '%s_trimmed.aln' % each_og
+
+        seq_file_out_handle = open(file_out_seq, 'w')
+        current_gnm_set = set()
+        for each_seq in SeqIO.parse(seq_file_in, 'fasta'):
+            seq_id = each_seq.id
+            gnm_id = '_'.join(seq_id.split('_')[:-1])
+            if seq_named_by_gnm is True:
+                gnm_id = seq_id
+            if gnm_id in gnm_to_process:
+                current_gnm_set.add(gnm_id)
+                seq_file_out_handle.write('>%s\n' % each_seq.id)
+                seq_file_out_handle.write('%s\n'  % each_seq.seq)
+        seq_file_out_handle.close()
+
+        cov_value = len(current_gnm_set)*100/len(gnm_to_process)
+        cov_value = float("{0:.2f}".format(cov_value))
+
+        if cov_value < cov_cutoff:
+            report_str = 'Ignored %s, contains proteins from %s (%s%s) genomes, < %s%s.' % (each_og, len(current_gnm_set), cov_value, '%', cov_cutoff, '%')
+            ignored_og_dict[each_og] = report_str
+            os.system('rm %s' % file_out_seq)
+        else:
+            # align, trim and iqtree
+            mafft_cmd  = 'mafft-einsi --thread %s --quiet %s.%s > %s'                                % (num_of_js_threads, each_og, fasta_file_ext, file_out_aln)
+            bmge_cmd   = 'java -jar %s -i %s -m %s -t AA -h %s -of %s'                               % (pwd_bmge_jar, file_out_aln, bmge_trim_model, bmge_entropy_score_cutoff, file_out_aln_trimmed)
+            iqtree_cmd = 'iqtree2 -s %s --seqtype AA -m %s -B 1000 --wbtl --bnni --prefix %s -T %s --quiet' % (file_out_aln_trimmed, iqtree_model, each_og, num_of_js_threads)
+            # Undinarchaeota illuminate DPANN phylogeny and the impact of gene transfer on archaeal evolution, settings: -m LG+G -bb 1000 -wbtl -bnni
+            cmds_in_one_line_txt_handle.write('%s; %s; %s\n' % (mafft_cmd, bmge_cmd, iqtree_cmd))
+    cmds_in_one_line_txt_handle.close()
+
+    # report ignored markers
+    if len(ignored_og_dict) > 0:
+        print('The following %s markers were ignored due to low genome coverage, see details in %s:' % (len(ignored_og_dict), ignored_marker_txt))
+        print('\n'.join(sorted(list(ignored_og_dict.keys()))))
+        ignored_marker_txt_handle = open(ignored_marker_txt, 'w')
+        for each_ignored_marker in sorted(list(ignored_og_dict.keys())):
+            ignored_marker_txt_handle.write(ignored_og_dict[each_ignored_marker] + '\n')
+        ignored_marker_txt_handle.close()
+
+    # report
+    print('You will need to execute the commands exported to the following file before moving to SplitScore2')
+    print(cmds_in_one_line_txt)
+    print('Done!')
+
+
+if __name__ == '__main__':
+
+    SplitScore1_parser = argparse.ArgumentParser()
+    SplitScore1_parser.add_argument('-i',                   required=True,                          help='orthologous gene sequence')
+    SplitScore1_parser.add_argument('-x',                   required=True,                          help='fasta file extension')
+    SplitScore1_parser.add_argument('-o',                   required=True,                          help='output directory')
+    SplitScore1_parser.add_argument('-u',                   required=False, default=None,           help='interested genomes, no file extension')
+    SplitScore1_parser.add_argument('-m',                   required=False, default='LG+G',         help='iqtree_model, default: LG+G')
+    SplitScore1_parser.add_argument('-c',                   required=False, type=int, default=75,   help='coverage cutoff, default: 75')
+    SplitScore1_parser.add_argument('-f',                   required=False, action="store_true",    help='force overwrite')
+    SplitScore1_parser.add_argument('-seq_named_by_gnm',    required=False, action="store_true",    help='named_by_gnm, specify if sequence named by gnm')
+    SplitScore1_parser.add_argument('-jst',                 required=False, type=int, default=1,    help='num of threads for iqtree2, default: 1')
+    args = vars(SplitScore1_parser.parse_args())
+    SplitScore1(args)

TreeSAK/SplitScore1OMA.py

@@ -0,0 +1,148 @@
+from __future__ import print_function
+import os
+import argparse
+from Bio import SeqIO
+
+
+SplitScore1OMA_usage = '''
+======================== SplitScore1OMA example commands ========================
+
+# SplitScore1
+TreeSAK SplitScore1OMA -i OrthologousGroups.txt -s OrthologousGroupsFasta -o step1_op_dir -t 6 -f
+TreeSAK SplitScore1OMA -i OrthologousGroups.txt -s OrthologousGroupsFasta -o step1_op_dir -t 6 -f -u interested_gnm.txt
+# Please ensure that all the commands in iqtree_cmds.txt have been executed before proceeding to step 2.
+
+=================================================================================
+'''
+
+
+def select_seq(seq_file, seq_id_list, output_file):
+    output_file_handle = open(output_file, 'w')
+    for seq_record in SeqIO.parse(seq_file, 'fasta'):
+        seq_id = seq_record.id
+        if seq_id in seq_id_list:
+            output_file_handle.write('>%s\n' % seq_id)
+            output_file_handle.write('%s\n' % str(seq_record.seq))
+    output_file_handle.close()
+
+
+def get_gene_tree(oma_op_txt, oma_op_fasta, interested_gnm_txt, cov_cutoff, oma_op_fasta_qualified, iqtree_model, num_of_js_threads, force_overwrite, get_gene_tree_cmd_txt):
+
+    # get the total number of genome
+    genome_id_set = set()
+    for each_group in open(oma_op_txt):
+        if not each_group.startswith('#'):
+            for each_gene in each_group.strip().split('\t')[1:]:
+                gnm_id = '_'.join(each_gene.split(':')[1].split(' ')[0].split('_')[:-1])
+                genome_id_set.add(gnm_id)
+
+    interested_gnm_set = set()
+    if interested_gnm_txt is not None:
+        for each_gnm in open(interested_gnm_txt):
+            interested_gnm_set.add(each_gnm.strip())
+    else:
+        interested_gnm_set = genome_id_set
+
+    # create output folder
+    if os.path.isdir(oma_op_fasta_qualified) is True:
+        if force_overwrite is True:
+            os.system('rm -r %s' % oma_op_fasta_qualified)
+        else:
+            print('%s already exist, program exited!' % oma_op_fasta_qualified)
+            exit()
+    os.system('mkdir %s' % oma_op_fasta_qualified)
+
+    # filter OMA output
+    qualified_grp_to_gene_dict = dict()
+    for each_group in open(oma_op_txt):
+        if not each_group.startswith('#'):
+            each_group_split  = each_group.strip().split('\t')
+            group_id          = each_group_split[0]
+            gene_list_by_gnm  = each_group_split[1:]
+            current_gene_list = [i.split(':')[1].split(' ')[0] for i in gene_list_by_gnm]
+            current_gnm_list_interested = []
+            current_gene_list_interested = []
+            for gene in current_gene_list:
+                gnm = '_'.join(gene.split('_')[:-1])
+                if gnm in interested_gnm_set:
+                    current_gnm_list_interested.append(gnm)
+                    current_gene_list_interested.append(gene)
+
+            current_cov = len(current_gnm_list_interested) * 100 / len(interested_gnm_set)
+            if current_cov >= cov_cutoff:
+                qualified_grp_to_gene_dict[group_id] = current_gene_list_interested
+
+    print('The number of orthologous groups with coverage >= %s is %s.' % (cov_cutoff, len(qualified_grp_to_gene_dict)))
+
+    # prepare commands for getting gene tree
+    get_gene_tree_cmd_txt_handle = open(get_gene_tree_cmd_txt, 'w')
+    for qualified_grp in sorted(list(qualified_grp_to_gene_dict.keys())):
+        group_id_only_num = qualified_grp.replace('OMA', '')
+        while group_id_only_num[0] == '0':
+            group_id_only_num = group_id_only_num[1:]
+
+        # define file name
+        og_id               = 'OG%s'                % group_id_only_num
+        pwd_seq_file_in     = '%s/%s.fa'            % (oma_op_fasta, og_id)
+        pwd_og_seq          = '%s/%s.fa'            % (oma_op_fasta_qualified, og_id)
+        pwd_og_aln          = '%s/%s.aln'           % (oma_op_fasta_qualified, og_id)
+        pwd_og_aln_trimmed  = '%s/%s_trimmed.aln'   % (oma_op_fasta_qualified, og_id)
+
+        # get sequence
+        if len(interested_gnm_set) == len(genome_id_set):
+            cp_cmd = 'cp %s %s' % (pwd_seq_file_in, pwd_og_seq)
+            os.system(cp_cmd)
+        else:
+            select_seq(pwd_seq_file_in, qualified_grp_to_gene_dict[qualified_grp], pwd_og_seq)
+
+        # align, trim and iqtree
+        mafft_cmd     = 'mafft-einsi --thread %s --quiet %s > %s'                                      % (num_of_js_threads, pwd_og_seq, pwd_og_aln)
+        trimal_cmd    = 'trimal -in %s -out %s -automated1'                                            % (pwd_og_aln, pwd_og_aln_trimmed)
+        iqtree_cmd    = 'iqtree2 -s %s --seqtype AA -m %s -T %s -B 1000 --quiet --wbtl --prefix %s/%s' % (pwd_og_aln_trimmed, iqtree_model, num_of_js_threads, oma_op_fasta_qualified, og_id)
+        cmds_one_line = '%s; %s; %s'                                                                   % (mafft_cmd, trimal_cmd, iqtree_cmd)
+        get_gene_tree_cmd_txt_handle.write(cmds_one_line.replace((oma_op_fasta_qualified + '/'), '') + '\n')
+    get_gene_tree_cmd_txt_handle.close()
+
+
+def SplitScore1OMA(args):
+
+    oma_op_txt              = args['i']
+    oma_op_fasta            = args['s']
+    interested_gnm_txt      = args['u']
+    iqtree_model            = args['m']
+    cov_cutoff              = args['c']
+    force_overwrite         = args['f']
+    num_of_js_threads       = args['jst']
+    step_1_op_dir           = args['o']
+
+    # define file name
+    qualified_og_dir    = '%s/qualified_OGs'    % step_1_op_dir
+    iqtree_cmds_txt     = '%s/iqtree_cmds.txt'  % step_1_op_dir
+
+    # create output folder
+    if os.path.isdir(step_1_op_dir) is True:
+        if force_overwrite is True:
+            os.system('rm -r %s' % step_1_op_dir)
+        else:
+            print('%s exist, program exited!' % step_1_op_dir)
+            exit()
+    os.mkdir(step_1_op_dir)
+    os.mkdir(qualified_og_dir)
+
+    # get get_gene_tree
+    get_gene_tree(oma_op_txt, oma_op_fasta, interested_gnm_txt, cov_cutoff, qualified_og_dir, iqtree_model, num_of_js_threads, force_overwrite, iqtree_cmds_txt)
+
+
+if __name__ == '__main__':
+
+    SplitScore1OMA_parser = argparse.ArgumentParser()
+    SplitScore1OMA_parser.add_argument('-i',   required=True,                        help='OrthologousGroups.txt, produced by OMA')
+    SplitScore1OMA_parser.add_argument('-s',   required=True,                        help='OrthologousGroupsFasta, produced by OMA')
+    SplitScore1OMA_parser.add_argument('-u',   required=False, default= None,        help='ID of interested genomes, no file extension')
+    SplitScore1OMA_parser.add_argument('-o',   required=True,                        help='output directory')
+    SplitScore1OMA_parser.add_argument('-m',   required=False, default='LG+G+I',     help='iqtree_model, default: LG+G+I')
+    SplitScore1OMA_parser.add_argument('-c',   required=False, type=int, default=80, help='coverage cutoff, default: 80')
+    SplitScore1OMA_parser.add_argument('-f',   required=False, action="store_true",  help='force overwrite')
+    SplitScore1OMA_parser.add_argument('-jst', required=False, type=int, default=1,  help='num of threads for inferring gene tree, default: 1')
+    args = vars(SplitScore1OMA_parser.parse_args())
+    SplitScore1OMA(args)