treesak 1.53.3__py3-none-any.whl

TreeSAK/pruneMSA.py

@@ -0,0 +1,63 @@
+import os
+import argparse
+
+
+pruneMSA_usage = '''
+==================== pruneMSA example commands ====================
+
+# Dependencies: perl and alignment_pruner.pl
+
+TreeSAK pruneMSA -i input_msa.fasta -c 10
+TreeSAK pruneMSA -i input_msa.fasta -c 5,10,20,30,40
+
+Note:
+1. This is a wrapper for alignment_pruner.pl (--chi2_prune mode) 
+2. For details: https://doi.org/10.1038/s41467-020-17408-w
+
+===================================================================
+'''
+
+
+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 pruneMSA(args):
+
+    msa_in              = args['i']
+    conserved_cutoffs   = args['c']
+
+    _, msa_path, msa_base, msa_ext = sep_path_basename_ext(msa_in)
+
+    current_file_path   = '/'.join(os.path.realpath(__file__).split('/')[:-1])
+    alignment_pruner_pl = '%s/alignment_pruner.pl'  % current_file_path
+    cutoff_list         = conserved_cutoffs.split(',')
+
+    op_file_list = []
+    for each_cutoff in cutoff_list:
+        cutoff_formatted = str(float(each_cutoff)/100)
+        current_msa_out     = '%s/%s_chi2p%s.%s'                        % (msa_path, msa_base, each_cutoff, msa_ext)
+        perl_cmd            = 'perl %s --file %s --chi2_prune f%s > %s' % (alignment_pruner_pl,   msa_in, cutoff_formatted, current_msa_out)
+        perl_cmd_to_print   = 'perl %s --file %s --chi2_prune f%s > %s' % ('alignment_pruner.pl', msa_in, cutoff_formatted, current_msa_out)
+        op_file_list.append(current_msa_out)
+        print(perl_cmd_to_print)
+        os.system(perl_cmd)
+
+    # report
+    print('Pruned MSA exported to:')
+    print('\n'.join(op_file_list))
+
+
+if __name__ == '__main__':
+
+    pruneMSA_parser = argparse.ArgumentParser()
+    pruneMSA_parser.add_argument('-i', required=True, help='input MSA file')
+    pruneMSA_parser.add_argument('-c', required=True, help='conservation cutoffs, comma separated')
+    args = vars(pruneMSA_parser.parse_args())
+    pruneMSA(args)

TreeSAK/recode.py

@@ -0,0 +1,73 @@
+import argparse
+from Bio import SeqIO
+
+
+recode_usage = '''
+============================ recode example commands ============================
+
+TreeSAK recode -i msa.fa -m s4 -o recoded_msa_SR4.fa
+TreeSAK recode -i msa.fa -m d4 -o recoded_msa_Dayhoff4.fa
+TreeSAK recode -i msa.fa -m d6 -o recoded_msa_Dayhoff6.fa
+
+Note:
+This script is modified based on the Recode_aa.py from Anja Spang.
+It was used to recode AA to SR4 (s4), Dayhoff4 (d4) or Dayhoff6 (d6) categories.  
+Please refer to https://doi.org/10.1038/s41467-020-17408-w for details.
+
+Recoding schemes
+1. Selenocysteine will be recoded to a gap.
+2. s4: A,G,N,P,S,T = A; C,H,W,Y = C; D,E,K,Q,R = G; F,I,L,M,V = T
+3. d4: A,G,P,S,T = A; D,E,N,Q = D; H,K,R = H; F,Y,W,I,L,M,V = F; C = ?
+4. d6: A,G,P,S,T = A; D,E,N,Q = D; H,K,R = H; F,Y,W = F; I,L,M,V = I; C = C
+
+=================================================================================
+'''
+
+
+def recode(args):
+
+    msa_in   = args['i']
+    msa_out  = args['o']
+    category = args['m']
+
+    DH6 = {'-': '-', 'A': 'A', 'G': 'A', 'P': 'A', 'S': 'A', 'T': 'A', 'D': 'D', 'E': 'D', 'N': 'D', 'Q': 'D', 'H': 'H', 'K': 'H', 'R': 'H', 'F': 'F', 'Y': 'F', 'W': 'F', 'I': 'I', 'L': 'I', 'M': 'I', 'V': 'I', 'C': 'C'}
+    DH4 = {'-': '-', 'A': 'A', 'G': 'A', 'P': 'A', 'S': 'A', 'T': 'A', 'D': 'D', 'E': 'D', 'N': 'D', 'Q': 'D', 'H': 'H', 'K': 'H', 'R': 'H', 'F': 'F', 'Y': 'F', 'W': 'F', 'I': 'F', 'L': 'F', 'M': 'F', 'V': 'F', 'C': '-'}
+    SR4 = {'-': '-', 'A': 'A', 'G': 'A', 'N': 'A', 'P': 'A', 'S': 'A', 'T': 'A', 'C': 'C', 'H': 'C', 'W': 'C', 'Y': 'C', 'D': 'G', 'E': 'G', 'K': 'G', 'Q': 'G', 'R': 'G', 'F': 'T', 'I': 'T', 'L': 'T', 'M': 'T', 'V': 'T'}
+
+    msa_out_handle = open(msa_out, 'w')
+    for seq_record in SeqIO.parse(msa_in, "fasta"):
+        header = str(seq_record.description).strip()
+        seq = str(seq_record.seq).strip()
+        new_seq = ''
+        for item in seq:
+            if category in ['D6', 'd6']:
+                if item in DH6:
+                    new_seq = new_seq + str(DH6.get(item))
+                else:
+                    new_seq = new_seq + str('-')
+            elif category in ['D4', 'd4']:
+                if item in DH4:
+                    new_seq = new_seq + str(DH4.get(item))
+                else:
+                    new_seq = new_seq + str('-')
+            elif category in ['S4', 's4']:
+                if item in SR4:
+                    new_seq = new_seq + str(SR4.get(item))
+                else:
+                    new_seq = new_seq + str('-')
+            else:
+                print('Please choose recoding scheme from d4, d6 and s4, program exited!')
+                exit()
+
+        msa_out_handle.write(">%s\n%s\n" % (header, new_seq))
+    msa_out_handle.close()
+
+
+if __name__ == '__main__':
+
+    recode_parser = argparse.ArgumentParser(description='Recode amino acids to Dayoff 4, Dayoff 6 or SR4 categories')
+    recode_parser.add_argument('-i', required=True,  help='input file')
+    recode_parser.add_argument('-m', required=True,  help='recoding scheme, choose from d4, d6 or s4')
+    recode_parser.add_argument('-o', required=True,  help='output file')
+    args = vars(recode_parser.parse_args())
+    recode(args)

TreeSAK/remove_bias.R

@@ -0,0 +1,112 @@
+library(Biostrings)
+options(digits=8)
+protein <- readAAStringSet("/lomi_home/wenxiu/RIS_virus/workdir/27_ortho_RIStree/ortho_tree/v1/tree_workdir/test.aln")
+outgroup <- c()
+protein <- protein[setdiff(names(protein),outgroup)]
+library(stringr)
+chi2_2 <- function(set,taxa_number,all_amino,amino_number){
+  O <- str_count(set[[taxa_number]],all_amino[amino_number])##某个物种中(已经除去一个位置的氨基酸之后)的全部氨基酸
+  E <- sum(str_count(set,all_amino[amino_number]))/sum(nchar(str_replace_all(set,"-","")))*nchar(str_replace_all(set[[taxa_number]],"-",""))###
+  return((O-E)^2/E)
+}
+
+chi2_1 <- function(set,taxa_number){
+  all_amino <- unique(strsplit(as.character(str_replace_all(set,"-","")),"")[[1]])
+  return(sum(sapply(1:length(all_amino),chi2_2,taxa_number=taxa_number,set=set,all_amino=all_amino)))
+}
+
+untrim <- sum(sapply(1:length(protein),chi2_1,set=protein))
+
+#13746.9
+
+trimmed_chi2 <- function(x){
+  library(Biostrings)
+  protein <- readAAStringSet("/lomi_home/wenxiu/RIS_virus/workdir/27_ortho_RIStree/ortho_tree/v1/tree_workdir/test.aln")
+  outgroup <- c()
+  protein <- protein[setdiff(names(protein),outgroup)]
+  library(stringr)
+  chi2_2 <- function(set,taxa_number,all_amino,amino_number){
+    O <- str_count(set[[taxa_number]],all_amino[amino_number])#某一个氨基酸(传进来的)在这个物种出现的频次
+    E <- sum(str_count(set,all_amino[amino_number]))/sum(nchar(str_replace_all(set,"-","")))*nchar(str_replace_all(set[[taxa_number]],"-",""))##氨基酸出现在配对中出现的的总次数/总氨基酸数*这个序列的氨基酸数
+    return((O-E)^2/E)
+  }
+  
+  chi2_1 <- function(set,taxa_number){###
+    all_amino <- unique(strsplit(as.character(str_replace_all(set,"-","")),"")[[1]])##除去补空位的,在该位置的所有氨基酸
+    return(sum(sapply(1:length(all_amino),chi2_2,taxa_number=taxa_number,set=set,all_amino=all_amino)))###对于其中的每一种氨基酸
+  }
+  
+  end_pos_1 <- x-1
+  start_pos_2 <- x+1
+  if (end_pos_1>=1 & start_pos_2<=width(protein)[1]){
+    trimmed_protein <- str_c(substr(protein,1,end_pos_1),substr(protein,start_pos_2,width(protein)[1]))###str_c元素对元素的合并列表
+    print(trimmed_protein)
+  }else if(start_pos_2<=width(protein)[1]){
+    trimmed_protein <- substr(protein,start_pos_2,width(protein)[1])#从start到全长
+  }else {
+    trimmed_protein <- substr(protein,1,end_pos_1)#从起始到end
+  }#到此截取了想要的氨基酸（所有序列的）
+  return(sum(sapply(1:length(trimmed_protein),chi2_1,set=trimmed_protein)))###对于每一个物种
+}
+
+
+library(parallel)
+
+cl<-makeCluster(48)
+
+trimmed_chi2_set <- parLapply(cl,1:width(protein)[1],trimmed_chi2)
+
+stopCluster(cl)
+
+save.image("/lomi_home/wenxiu/RIS_virus/workdir/27_ortho_RIStree/ortho_tree/v1/tree_workdir/test_removed_bias.aln")
+
+
+###删除对应的位点
+trim_site<-function(percentage){
+  protein1<-protein
+  realnum=c()
+  for (i in 1:length(trimmed_chi2_set)) {
+    realnum= c(realnum,trimmed_chi2_set[[i]])
+  }
+  #realnum=abs(realnum-untrim)
+  print(percentage*length(realnum))
+  kafang<-realnum[order(realnum,decreasing = F)[round(percentage*length(realnum))]]
+  site_to_be_removed=which(realnum<=kafang)
+  site<-IRanges(start=site_to_be_removed, end=site_to_be_removed, width=1)
+  protein
+  replaceAt(protein1,site,'')
+  protein1<-replaceAt(protein1,site,'')
+  substr(protein1[1],start=1,stop=width(protein1)[1])
+  #10：5984
+  #print(paste0('~/removed_bias_dataset3_concatenate_',percentage,'.txt'))
+  writeXStringSet(protein1,paste0('/lomi_home/wenxiu/RIS_virus/workdir/27_ortho_RIStree/ortho_tree/v1/tree_workdir/test_removed_bias.trimmed.concat_',percentage*100,'%.fas'))
+}
+
+  
+for (cutoff in c(0.05,0.1,0.2,0.3,0.4,0.6,0.8,0.9)) {
+  trim_site(cutoff)
+}
+
+
+
+
+
+
+
+
+###先分成两份，再取5%
+# protein1<-protein
+# realnum=c()
+# for (i in 1:length(trimmed_chi2_set)) {
+#   realnum= c(realnum,trimmed_chi2_set[[i]])
+# }
+# realnumplus<-realnum[which(realnum-untrim>0)]
+# realnummiuis<-realnum[which(realnum-untrim<0)]
+# plus<-realnumplus[order(realnumplus,decreasing = T)[1:(0.1*length(realnumplus)+1)]]
+# miuis<-realnummiuis[order(realnummiuis,decreasing = F)[1:(0.1*length(realnummiuis)+1)]]
+# site_to_be_removed=which(realnum %in% c(plus,miuis))
+# site<-IRanges(start=site_to_be_removed, end=site_to_be_removed, width=1)
+# protein
+# replaceAt(protein1,site,'')
+# protein1<-replaceAt(protein1,site,'')
+# substr(protein1[1],start=1,stop=5984)

TreeSAK/rename_leaves.py

@@ -0,0 +1,78 @@
+import os
+import argparse
+from ete3 import Tree
+
+
+rename_leaves_usage = '''
+==================== rename_leaves example commands ====================
+
+TreeSAK rename_leaves -i input.tree -r rename.txt -o output.tree
+
+# format of rename.txt (tab separated)
+leaf_1  leaf_1_new_name
+leaf_2  leaf_2_new_name
+
+========================================================================
+'''
+
+
+def rename_leaves(args):
+
+    tree_file_in  = args['i']
+    rename_file   = args['r']
+    tree_format   = args['f']
+    tree_file_out = args['o']
+
+    if os.path.isfile(tree_file_in) is False:
+        print('Tree file not found, program exited!')
+        exit()
+
+    if os.path.isfile(rename_file) is False:
+        print('Rename file not found, program exited!')
+        exit()
+
+    mag_rename_dict = {}
+    for each_mag in open(rename_file):
+        each_mag_split = each_mag.strip().split('\t')
+        before_rename = each_mag_split[0]
+        after_rename = each_mag_split[1]
+        mag_rename_dict[before_rename] = after_rename
+
+    t = Tree(tree_file_in, format=tree_format)
+
+    input_tree_leaf_name_list = []
+    for leaf in t:
+        input_tree_leaf_name_list.append(leaf.name)
+
+    leaves_with_new_name = 0
+    leaves_without_new_name = 0
+    for each_raw_name in input_tree_leaf_name_list:
+        if each_raw_name in mag_rename_dict:
+            leaves_with_new_name += 1
+        else:
+            leaves_without_new_name += 1
+
+    if leaves_with_new_name == 0:
+        print('No leaf on input tree found in rename file, please double check!')
+        exit()
+    elif leaves_without_new_name > 0:
+        print('%s of the %s leaves in input tree were found in the rename file, the rests unchanged.' % (leaves_with_new_name, len(input_tree_leaf_name_list)))
+
+    for leaf in t:
+        leaf_name_new = mag_rename_dict.get(leaf.name, leaf.name)
+        leaf.name = leaf_name_new
+    t.write(format=tree_format, outfile=tree_file_out)
+
+    print('Done!')
+
+
+if __name__ == '__main__':
+
+    rename_leaves_parser = argparse.ArgumentParser()
+    rename_leaves_parser.add_argument('-i',    required=True,                       help='input tree')
+    rename_leaves_parser.add_argument('-r',    required=True,                       help='rename file')
+    rename_leaves_parser.add_argument('-f',    required=False, default=1, type=int, help='tree format, default: 1')
+    rename_leaves_parser.add_argument('-o',    required=True,                       help='output tree')
+
+    args = vars(rename_leaves_parser.parse_args())
+    rename_leaves(args)

TreeSAK/replace_clade.py

@@ -0,0 +1,55 @@
+import argparse
+from ete3 import Tree
+
+
+replace_clade_usage = '''
+===================== replace_clade example commands =====================
+
+TreeSAK replace_clade -m main.tree -s sub.tree -l leaves.txt -o out.tree
+
+==========================================================================
+'''
+
+
+def replace_clade(args):
+
+    main_tree_file  = args['m']
+    sub_tree_file   = args['s']
+    leaf_txt        = args['l']
+    tree_out        = args['o']
+    tree_out_fmt    = args['of']
+
+    # read in subtree
+    sub_tre = Tree(sub_tree_file, quoted_node_names=True, format=1)
+
+    # read in leaves
+    leaf_list = []
+    for each_leaf in open(leaf_txt):
+        leaf_list.append(each_leaf.strip())
+
+    # read in main tree
+    main_tre = Tree(main_tree_file, quoted_node_names=True, format=1)
+
+    # remove clades
+    lca = main_tre.get_common_ancestor(leaf_list)
+
+    lca_p = lca.up
+    lca_p.remove_child(lca)
+    lca_p.add_child(sub_tre)
+
+    # write out updated tree
+    main_tre.write(outfile=tree_out, format=tree_out_fmt)
+
+
+
+
+if __name__ == '__main__':
+
+    replace_clade_parser = argparse.ArgumentParser()
+    replace_clade_parser.add_argument('-m',   required=True,                        help='main tree file')
+    replace_clade_parser.add_argument('-s',   required=True,                        help='subtree file')
+    replace_clade_parser.add_argument('-l',   required=True,                        help='leaves on main tree to be replaced')
+    replace_clade_parser.add_argument('-o',   required=True,                        help='output tree')
+    replace_clade_parser.add_argument('-of',  required=False, default=9, type=int,  help='output tree format, default is 9')
+    args = vars(replace_clade_parser.parse_args())
+    replace_clade(args)

TreeSAK/root_with_out_group.py

@@ -0,0 +1,84 @@
+from ete3 import Tree
+
+
+def root_with_out_group(tree_file, out_group_txt, tree_file_rooted):
+
+    out_group_set = set()
+    for each_og in open(out_group_txt):
+        out_group_set.add(each_og.strip())
+
+    tre = Tree(tree_file, format=1)
+    out_group_lca = tre.get_common_ancestor(out_group_set)
+    tre.set_outgroup(out_group_lca)
+    tre.write(outfile=tree_file_rooted)
+
+
+def replace_clades(main_tree, sub_tree, tree_out):
+
+    # read in sub tree
+    tre_sub = Tree(sub_tree, format=1)
+
+    # get all leaves in sub tree
+    subtree_leaf_name_list = tre_sub.get_leaf_names()
+
+    # read in main tree
+    tre_main = Tree(main_tree)
+
+    # remove clades
+    lca = tre_main.get_common_ancestor(subtree_leaf_name_list)
+
+    if len(lca.get_leaf_names()) != len(subtree_leaf_name_list):
+        print('LCA of subtree leaves in main tree contain extra leaves, program exited!')
+        exit()
+
+    lca_p = lca.up
+    lca_p.remove_child(lca)
+    lca_p.add_child(tre_sub)
+
+    # write out updated tree
+    tre_main.write(outfile=tree_out, format=8)
+
+
+tree_file                           = '/Users/songweizhi/Desktop/777/PA_75_DeltaLL_50_raw.treefile'
+out_group_txt                       = '/Users/songweizhi/Desktop/777/out_group.txt'
+tree_file_rooted                    = '/Users/songweizhi/Desktop/777/PA_75_DeltaLL_50_rooted.treefile'
+eu_tree                             = '/Users/songweizhi/Desktop/777/27.nwk'
+rooted_tree_with_time_constraints   = '/Users/songweizhi/Desktop/777/PA_75_DeltaLL_50_rooted_with_time_constraints.treefile'
+
+
+root_with_out_group(tree_file, out_group_txt, tree_file_rooted)
+replace_clades(tree_file_rooted, eu_tree, rooted_tree_with_time_constraints)
+
+
+root_with_out_group_from_tianhua = '''
+from ete3 import Tree
+tpath = ''
+nodes = ['F7','B7_3','A7']
+tre = Tree(tpath)
+tre2 = tre.copy()
+lca = tre.get_common_ancestor(nodes)
+lca_leaves = lca.get_leaf_names()
+# intersect = set(lca_leaves).intersection(set(nodes))
+ratio = len(nodes)/len(lca_leaves)
+if ratio > 0.5:
+    tre2.set_outgroup(lca)  # inplace
+    tre2.write(outfile='')
+else:
+    print(ratio)
+'''
+
+
+replace_clade_from_tianhua ='''
+## replace clade with a new tree    
+tre2 = Tree(tpath2,format=3)  
+nodes1 = []
+tre = Tree(tpath)
+lca = tre.get_common_ancestor(nodes1)
+lca_p = lca.up
+lca_p.remove_child(lca)
+if len(tre2.children)==1:
+    lca_p.add_child(tre2.children[0])
+else:
+    lca_p.add_child(tre2)
+tre.write(outfile='',format='')
+'''