From 942f7d60f4d10f6b38bd3b9ea30132ef0976bf47 Mon Sep 17 00:00:00 2001
From: sauloal <sauloal@yahoo.com.br>
Date: Mon, 2 Mar 2015 19:04:55 +0100
Subject: [PATCH] complete phylogeny
---
filler.py | 133 +++++++++++++++++++++++++++++++++++++++---------------
1 file changed, 96 insertions(+), 37 deletions(-)
diff --git a/filler.py b/filler.py
index 1e4b1e3..5698278 100755
--- a/filler.py
+++ b/filler.py
@@ -2,17 +2,19 @@
import os
import sys
+from collections import defaultdict
import parser_SQL_struct
+col_names = "species|species subgroup|species group|subgenus|genus|subtribe|tribe|subfamily|family|superfamily|parvorder|infraorder|suborder|order|superorder|infraclass|subclass|class|superclass|subphylum|phylum|subkingdom|kingdom|superkingdom|root"
+
class csv_holder(object):
- def __init__(self, incsv, col_names):
+ def __init__(self, incsv):
self.incsv = incsv
- self.col_names = col_names
- self.num_levels = len(col_names)
self.names = []
self.data = {}
+ self.col_names = col_names.split("|")
self.read_csv()
def read_csv(self):
@@ -38,10 +40,14 @@ class csv_holder(object):
self.names.append( name )
- self.data[ name ] = [None]*self.num_levels
+ self.data[ name ] = [None]*len(self.col_names)
+
+ #if len(self.data) > 15:
+ # break
def save(self, filename):
print "saving to %s" % filename
+
with open(filename, 'w') as fhd:
#print self.col_names
fhd.write("name\t" + "\t".join(self.col_names) + "\n")
@@ -50,12 +56,17 @@ class csv_holder(object):
#print data
cols = name + "\t" + "\t".join([str(d) if d is not None else "" for d in data])
fhd.write( cols + "\n" )
-
+
+
class filler(object):
def __init__(self, csv, querier):
self.csv = csv
self.querier = querier
+
+ self.csv.col_names.insert(0, "tax_id" )
+ self.csv.col_names.insert(0, "division")
+
self.get_ids()
self.get_taxonomy()
@@ -66,20 +77,23 @@ class filler(object):
if len(tax_ids) == 0:
print "Species '%s' does not exists" % name
sys.exit(1)
+
elif len(tax_ids) > 1:
print "More than one instance of %s" % name
sys.exit(1)
+
else:
tax_id = tax_ids[0]
print "adding species %-30s id %7d" % (name, tax_id)
- self.csv.data[name][1] = tax_id
- #break
-
+ self.csv.data[name].insert(0, tax_id )
+
def get_taxonomy(self):
+ datas = {}
+
for name in sorted(self.csv.data):
- tax_id = self.csv.data[name][1]
+ tax_id = self.csv.data[name][0]
- print "getting taxonomy of '%s' id %d" % ( name, tax_id )
+ print "getting taxonomy of %-30s id %d" % ( "'%s'"%name, tax_id )
node = get_node(tax_id)
#print " node", node
@@ -87,41 +101,84 @@ class filler(object):
division_id = node.division_id
division_name = get_division_name(division_id)
- print " division_id", division_id, "division_name", division_name
+ datas[name] = data
-
+ self.csv.data[name].insert(0, division_name)
+
+ for name, data in datas.items():
for d in data:
- p_rank = d[0]
- p_tax_id = d[1]
- p_name = get_name_from_tax_id(p_tax_id)[0]
- d[3] = p_name
- print " parent rank %-15s id %7d name %-30s" % (p_rank, p_tax_id, p_name)
-
+ p_rank = d[0]
+ p_tax_id = d[1]
+ p_name = get_name_from_tax_id(p_tax_id )[0] if p_tax_id is not None else "None"
+ d[2] = p_name
+
+ p_offspring_rank = d[3]
+ p_offspring_tax_id = d[4]
+ p_offspring_name = get_name_from_tax_id(p_offspring_tax_id)[0] if p_offspring_tax_id is not None else "None"
+ d[5] = p_offspring_name
+
+ p_parent_rank = d[6]
+ p_parent_tax_id = d[7]
+ p_parent_name = get_name_from_tax_id(p_parent_tax_id )[0] if p_parent_tax_id is not None else "None"
+ d[8] = p_parent_name
+
+ print " parent rank %-17s id %s name %-30s off (%-17s %s %-30s) par (%-17s %s %-30s)" % \
+ (p_rank , "%7d"%p_tax_id if p_tax_id is not None else " None", p_name,
+ p_offspring_rank, "%7d"%p_offspring_tax_id if p_offspring_tax_id is not None else " None", p_offspring_name,
+ p_parent_rank , "%7d"%p_parent_tax_id if p_parent_tax_id is not None else " None", p_parent_name)
+
+ if p_rank not in self.csv.col_names:
+ print "unknown rank %s" % p_rank
+ sys.exit(1)
+
if p_rank in self.csv.col_names:
p_pos = self.csv.col_names.index(p_rank)
self.csv.data[name][p_pos] = p_name
-
- self.csv.data[name][0] = division_name
- #print self.csv.data[name]
-
- #break
+ print
+
+
+
def parse_node(node):
data = []
- parent = node.parent
- data.append( [node.rank, node.tax_id, parent.parent.tax_id, None] )
#print " ", data[-1]
- while parent.tax_id != 1:
- data.append( [parent.rank, parent.tax_id, parent.parent.tax_id, None] )
- #print " ", data[-1]
- parent = parent.parent
-
- data.append( [parent.rank, parent.tax_id, parent.parent.tax_id, None] )
+
+ prev_rank = None
+ current_rank = node.rank
+ next_rank = node.parent.rank
+
+ prev_tax_id = None
+ current_tax_id = node.tax_id
+ next_tax_id = node.parent.tax_id
+
+ current = node
+
+ while current_tax_id != 1:
+ #print current
+ if current_rank != "no rank":
+ while next_rank == "no rank" and next_tax_id != 1:
+ current = current.parent
+ next_rank = current.parent.rank
+ next_tax_id = current.parent.tax_id
+
+ data.append( [current_rank, current_tax_id, None, prev_rank, prev_tax_id, None, next_rank, next_tax_id, None] )
+
+ prev_rank = current_rank
+ prev_tax_id = current_tax_id
+
+ current = current.parent
+ current_rank = current.rank
+ current_tax_id = current.tax_id
+ next_rank = current.parent.rank
+ next_tax_id = current.parent.tax_id
+
+ current_rank = "root"
+ data[-1][6] = current_rank
+ data.append( [current_rank, current_tax_id, None, prev_rank, prev_tax_id, None, None, None, None] )
- #print " ", data[-1]
return data
def get_ranks():
@@ -129,7 +186,7 @@ def get_ranks():
ranks = session.query(field).distinct().all()
ranks = [r.rank for r in ranks]
ranks.sort()
- ranks.remove("no rank")
+ #ranks.remove("no rank")
return ranks
def query_name(name):
@@ -190,10 +247,12 @@ def main(args):
elif cmd == "fill":
incsv = args[1]
- ranks = get_ranks()
- ranks.insert(0, "tax_id" )
- ranks.insert(0, "division")
- holder = csv_holder(incsv, ranks)
+ #ranks = get_ranks()
+ #ranks = []
+ #ranks.insert(0, "tax_id" )
+ #ranks.insert(0, "division")
+ #holder = csv_holder(incsv, ranks)
+ holder = csv_holder(incsv)
fill = filler(holder, get_tax_id_from_name)
holder.save(incsv + '.filled.csv')
--
GitLab