def get_btree(treestr):
if treestr.count(':') == 1: # one-leaf tree
name, lengthstr = treestr.strip().rstrip(';').split(':')
tree = OneLeafTree(name, float(lengthstr))
else:
tree = baltic.tree()
baltic.make_tree(treestr, tree, verbose=False)
tree.traverse_tree()
return tree
), 'Expected number of tips: %s\nNumber of tips found: %s' % (
tipNum, len(tips)
) ## check that correct numbers of tips have been parsed
################################################################################### start analysing trees
cerberus = re.match(
'tree\sSTATE\_([0-9]+).+\[\&R\]\s', line
) ## search for crud at the beginning of the line that's not a tree string
if cerberus is not None: ## tree identified
################################################################# at state 0 - create the header for the output file and read the tree (in case the output log file requires information encoded in the tree)
if treecount == 0: ## At tree state 0 insert header into output file
ll = bt.tree() ## empty tree object
start = len(cerberus.group()
) ## index of where tree string starts in the line
treestring = str(line[start:]) ## grab tree string
bt.make_tree(treestring, ll) ## read tree string
if lower == 0 and upper == np.inf: ## only add a header if not doing a chunk
outfile.write('state') ## begin the output log file
########################################### add header to output log file
if 'treeLength' in analyses:
outfile.write('\ttreeLength')
###########################################
if 'RC' in analyses:
outfile.write('\tN\tS\tuN\tuS\tdNdS')
###########################################
if 'tmrcas' in analyses:
tmrcas = {
'A': [],
'B': [],
'C': []
} ## dict of clade names
cerberus = re.search(
'dimensions ntax=([0-9]+);',
l.lower()) ## check how many tips there are supposed to be
if cerberus is not None:
tipNum = int(cerberus.group(1))
#####################
cerberus = re.search(
'tree TREE([0-9]+) = \[&R\]',
l) ## search for beginning of tree string in BEAST format
if cerberus is not None:
treeString_start = l.index(
'(') ## tree string starts where the first '(' is in the line
ll = bt.tree() ## new instance of tree
bt.make_tree(
l[treeString_start:], ll
) ## send tree string to make_tree function, provide an empty tree object
#####################
if tipFlag == True:
cerberus = re.search(
'([0-9]+) ([A-Za-z\-\_\/\.\'0-9 \|?]+)', l
) ## look for tip name map, where each tip is given an integer to represent it in tree
if cerberus is not None:
tips[cerberus.group(1)] = cerberus.group(2).strip(
"'"
) ## if you give tips an integer (in the form of a string), it will return the full name of the tip
elif ';' not in l: ## something's wrong - nothing that matches the tip regex is being captured where it should be in the file
print 'tip not captured by regex:', l.replace('\t', '')
if 'translate' in l.lower(): ## start looking for tips
cerberus=re.search('([0-9]+) ([\'\"A-Za-z0-9\?\|\-\_\.\/]+)',line)
tips[cerberus.group(1)]=cerberus.group(2).strip("'")
if 'tree STATE_' in line and plate==True: ## starting actual analysis
plate=False
assert (tipNum == len(tips)),'Expected number of tips: %s\nNumber of tips found: %s'%(tipNum,len(tips)) ## check that correct numbers of tips have been parsed
################################################################################### start analysing trees
cerberus=re.match('tree\sSTATE\_([0-9]+).+\[\&R\]\s',line) ## search for crud at the beginning of the line that's not a tree string
if cerberus is not None: ## tree identified
################################################################# at state 0 - create the header for the output file and read the tree (in case the output log file requires information encoded in the tree)
if treecount==0: ## At tree state 0 insert header into output file
ll=bt.tree() ## empty tree object
start=len(cerberus.group()) ## index of where tree string starts in the line
treestring=str(line[start:]) ## grab tree string
bt.make_tree(treestring,ll) ## read tree string
if lower==0 and upper==np.inf: ## only add a header if not doing a chunk
outfile.write('state') ## begin the output log file
########################################### add header to output log file
if 'treeLength' in analyses:
outfile.write('\ttreeLength')
###########################################
if 'RC' in analyses:
outfile.write('\tN\tS\tuN\tuS\tdNdS')
###########################################
if 'tmrcas' in analyses:
tmrcas={'A':[],'B':[],'C':[]} ## dict of clade names
ll.renameTips(tips)
for k in ll.Objects: ## iterate over branches
if isinstance(k,bt.leaf): ## only interested in tips
if 'A' in k.name: ## if name of tip satisfies condition
try:
treefiles = sorted(
[t for t in glob('*bestTree*') if t.split('_')[-1] in protein_list],
key=lambda t: protein_list.index(t.split('_')[-1]))
assert len(treefiles) == len(protein_list)
except ValueError as e: # For now, require that we can match all tree files to a protein
print 'Oops! Plotting without building trees? Make sure your trees are named like `whatever_protein` and you passed a list of matching `protein` names to `-p` to set the order of trees.\n\n'
raise e
except AssertionError as e:
print 'ERROR: Missing tree files. Looked for trees for these proteins:\n', protein_list, '\n\n'
raise e
trees = {}
for i, t in enumerate(treefiles):
treestring, treeobject = open(t, 'r').readline().strip(), bt.tree()
bt.make_tree(treestring, treeobject)
treeobject.treeStats() ## initial traversal, checks for stats
treeobject.sortBranches(
) ## traverses tree, sorts branches, draws tree (sets plotting coordinates)
trees[i] = treeobject
for i in range(1, len(treefiles)):
print 'Untangling tree number %d' % i
untangle(trees[i - 1], trees[i])
################
## Plot Genome Map
################
if proteins == None and reference != None: # If we didn't parse proteins earlier, but have the reference sequence, do so now.
proteins, reference_seq = load_reference(reference)
reference_seq = str(reference_seq.seq)