if [ch.traits[traitName]
for ch in k.children].count(kloc) > 0:
proceed = True
#print k.index,k.parent.index,k.traits,k.parent.traits,k.traits[traitName]
if proceed == True: ## if at least one valid tip and no hanging nodes
subtree = copy.deepcopy(
ll.traverseWithinTrait(k, traitName))
subtree_leaves = [
x.name for x in subtree
if isinstance(x, bt.leaf)
]
if len(subtree_leaves) > 0:
mostRecentTip = max([
bt.decimalDate(
x.strip("'").split('|')[-1])
for x in subtree_leaves
])
while sum(
[
len(nd.children) - sum([
1 if ch in subtree else 0
for ch in nd.children
]) for nd in subtree if isinstance(
nd, bt.node) and nd.index != 'Root'
]
) > 0: ## keep removing nodes as long as there are nodes with children that are not entirely within subtree
for nd in sorted(
[
q for q in subtree
if isinstance(q, bt.node)
if 'translate' in l.lower(): ## start looking for tips
tipFlag = True
if ';' in l: ## stop looking for tips
tipFlag = False
print "Number of objects found in tree string: %d" % (len(ll.Objects))
## rename tips, find the highest tip (in absolute time) in the tree
if len(
tips
) == 0: ## use this if tip names in the string are already the final format
for k in ll.Objects:
if isinstance(k, bt.leaf):
k.name = k.numName
highestTip = max([
bt.decimalDate(x.name.strip("'").split('|')[-1].replace("_", "-"),
variable=True) for x in ll.Objects
if isinstance(x, bt.leaf)
])
else: ## there's a tip name map at the beginning, so translate the names
ll.renameTips(tips) ## give each tip a name
highestTip = max([
bt.decimalDate(x.strip("'").split('|')[-1].replace("_", "-"),
variable=True) for x in tips.values()
])
ll.sortBranches()
ll.setAbsoluteTime(highestTip)
lowestTip = min([x.absoluteTime for x in ll.Objects if x != ll.root])
timeTraversed = highestTip - lowestTip
tips) ## Rename tips so their name refers to sequence name
else:
for k in ll.Objects:
if isinstance(k, leaf):
k.name = k.numName ## otherwise every tip gets a name that's the same as tree string names
#### calibration
dateCerberus = re.compile(
tformat) ## search pattern + brackets on actual calendar date
if calibration == True: ## Calibrate tree so everything has a known position in actual time
tipDatesRaw = [
dateCerberus.search(x).group(1) for x in tips.values()
]
tipDates = []
for tip_date in tipDatesRaw:
tipDates.append(
bt.decimalDate(tip_date, fmt=dformat, variable=True))
maxDate = max(tipDates) ## identify most recent tip
ll.setAbsoluteTime(maxDate)
outfile.write('%s' % cerberus.group(1)
) ## write MCMC state number to output log file
################################################################################
if 'treeLength' in analyses:
treeL = sum([k.length for k in ll.Objects]) ## do analysis
outfile.write('\t%s' % (treeL)) ## output to file
###################################################
if 'RC' in analyses: ## 'RC' was queued as an analysis
Ns = [] ## empty list
Ss = []
uNs = []
uSs = []
for k in ll.Objects: ## iterate over branch objects in the tree
bt.make_tree(treestring,ll) ## pass it to make_tree function
ll.traverse_tree() ## Traverse the tree - sets the height of each object in the tree
#### renaming tips
if len(tips)>0:
ll.renameTips(tips) ## Rename tips so their name refers to sequence name
else:
for k in ll.Objects:
if isinstance(k,leaf):
k.name=k.numName ## otherwise every tip gets a name that's the same as tree string names
#### calibration
dateCerberus=re.compile(tformat) ## search pattern + brackets on actual calendar date
if calibration==True: ## Calibrate tree so everything has a known position in actual time
tipDatesRaw=[dateCerberus.search(x).group(1) for x in tips.values()]
tipDates=[]
for tip_date in tipDatesRaw:
tipDates.append(bt.decimalDate(tip_date,fmt=dformat,variable=True))
maxDate=max(tipDates) ## identify most recent tip
ll.setAbsoluteTime(maxDate)
outfile.write('%s'%cerberus.group(1)) ## write MCMC state number to output log file
################################################################################
if 'treeLength' in analyses:
treeL=sum([k.length for k in ll.Objects]) ## do analysis
outfile.write('\t%s'%(treeL)) ## output to file
###################################################
if 'RC' in analyses: ## 'RC' was queued as an analysis
Ns=[] ## empty list
Ss=[]
uNs=[]
uSs=[]
for k in ll.Objects: ## iterate over branch objects in the tree
if k.traits.has_key('N'): ## if branch has a trait labelled "N"...
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
tipFlag=True
if ';' in l: ## stop looking for tips
tipFlag=False
print "Number of objects found in tree string: %d"%(len(ll.Objects))
## rename tips, find the highest tip (in absolute time) in the tree
if len(tips)==0: ## use this if tip names in the string are already the final format
for k in ll.Objects:
if isinstance(k,bt.leaf):
k.name=k.numName
highestTip=max([bt.decimalDate(x.name.strip("'").split('|')[-1].replace("_","-"),variable=True) for x in ll.Objects if isinstance(x,bt.leaf)])
else: ## there's a tip name map at the beginning, so translate the names
ll.renameTips(tips) ## give each tip a name
highestTip=max([bt.decimalDate(x.strip("'").split('|')[-1].replace("_","-"),variable=True) for x in tips.values()])
ll.sortBranches()
ll.setAbsoluteTime(highestTip)
lowestTip=min([x.absoluteTime for x in ll.Objects if x != ll.root])
timeTraversed = highestTip - lowestTip
coords = pd.read_csv("data/coordinates.csv", names=["Country", "lat", "lng", "Continent"])
coords["Country_NA_EARTH"] = coords["Country"]
# Match country names in coordinates.csv with names in shape file