# *** setquery.py can be used to search presence of OGs for 3+ groups. This can be used to link OGs to particular character traits. ***
# Now modified to enable the use of different sp. code dictionaries and group lists in analysis, determined by sys.argv[1] and sys.argv[2]..
import group
import sys
# sys.argv[1] must equal groups, alt_groups or alt_groups_18.
# sys.argv[2] must equal codes, alt_codes, or alt_codes_18.
# function_mappings and select_function() function are defined to map the sys.argvs from strings to actual functions.
# Comma-separated returns allow multiple returns from the same function (only just realised this).
function_mappings = {
'groups': group.groups(),
'alt_groups': group.alt_groups(),
'alt_groups_18': group.alt_groups_18(),
'codes': group.codes(),
'alt_codes': group.alt_codes(),
'alt_codes_18': group.alt_codes_18()
}
def select_function():
while True:
try:
return function_mappings[sys.argv[1]], function_mappings[
sys.argv[2]]
except KeyError:
print('Invalid function, try again.')
# select_function is called with the relevant group list and codes - these should correspond else the program may fail.
group_list, sp_codes = select_function()
# Another alternative to the original find_group.py, iterates over the divided SAR and Haptista subgroups. # New script for the extra groups is created as an alternative to iterating over every group again. import re import glob import group codes = group.alt_codes_18() new_groups = ['Alveolata', 'Centrohelids', 'Haptophyta', 'Rhizaria', 'Stramenopiles'] all_groups = group.alt_groups_18() sorted_all_groups = sorted(all_groups) all_sets = [] for eugroup in new_groups: for ogroup in sorted_all_groups: query = [eugroup, ogroup] if eugroup != ogroup: group_set = set(query) else: group_set = eugroup if group_set not in all_sets: # Appends to all sets to ensure no duplicate analyses. all_sets.append(group_set) if isinstance(group_set, set): group_set = sorted(group_set) # Reverts data to a sorted list so that names are alphabetical. group.find_group(group_set, codes) # 2nd argument of find_group() can now be configured depending on which code set is needed.
]
correct_order_15 = [
"Telonemids", "Haptista", "SAR", "Atwista", "Archaeplastida",
"Ancyromonadida", "Obazoa", "Discoba", "Collodictyonids", "Cryptista",
"Amoebozoa", "Metamonads", "Hemimastigophora", "Apusomonada",
"Malawimonadidae"
]
correct_order_includingown_18 = [
"Alveolata", "Stramenopiles", "Archaeplastida", "Obazoa", "Telonemids",
"Centrohelids", "Ancyromonadida", "Discoba", "Rhizaria", "Cryptista",
"Atwista", "Haptophyta", "Collodictyonids", "Amoebozoa", "Metamonads",
"Hemimastigophora", "Apusomonada", "Malawimonadidae"
]
# Choose the necessary group list from the group module.
alt_groups = sorted(group.alt_groups_18())
group_names = alt_groups
data = []
for name in group_names:
to_parse = glob.glob(
"/mnt/c/Users/scamb/Documents/uob_msc/Genome_data/OG_arb-fal/new_outputs/*.txt"
)
for file in to_parse:
filename = re.search(r"([A-Z]\w*)_([A-Z]\w*)_output.txt$", file)
if filename:
if filename.group(
1
) not in group_names: # Skips 'non-target' files in new_outputs.
pass
elif filename.group(
# This script finds all the OGs of each group, regardless of exclusivity.
# Outputs are stored in total_genome directory.
import glob
import group
import re
codes = group.alt_codes_18()
group_names = group.alt_groups_18() # alt_groups_18() has eukaryote groups SAR and Haptista, as well as 'Other', split into their respective subgroups.
for name in group_names:
to_parse = glob.glob("*.fal")
filename = "%s_allOGs.txt" % name
genome = open(filename, "w")
i = 0
for file in to_parse:
groups_present = []
with open(file) as f:
for line in f:
if line.startswith(">"):
fields = re.split("_", line)
species_code = fields[0][1:]
for code in codes:
group = codes[species_code]
if group not in groups_present:
groups_present.append(group)
if name in groups_present: # Functions in the same way as findgroup, only without the exclusive set matching.
genomeWrite = genome.write(f'{file}\n')
i += 1
genomeWrite = genome.write(f'Total gene families in group: {i}')
genome.close()