gsrap 0.8.2__py3-none-any.whl → 0.9.0__py3-none-any.whl

gsrap/.ipynb_checkpoints/__init__-checkpoint.py

@@ -75,12 +75,14 @@ def main():
     parsedb_parser.add_argument("-z", "--initialize", metavar='', type=str, default='-', help="Initialize the universe on the provided medium. By default, the first medium in --media is used. Use 'none' to avoid initialization.")
     parsedb_parser.add_argument("--precursors", action='store_true', help="Verify biosynthesis of biomass precursors and show blocked ones.")
     parsedb_parser.add_argument("--biosynth", action='store_true', help="Check biosynthesis of all metabolites and detect dead-ends.")
+    parsedb_parser.add_argument("-t", "--taxon", metavar='', type=str, default='-', help="High-level taxon of interest. If provided, it must follow the syntax '{level}:{name}', where {level} is 'kingdom' or 'phylum'.")
     parsedb_parser.add_argument("-e", "--eggnog", nargs='+', metavar='', type=str, default='-', help="Path to the optional eggnog-mapper annotation table(s).")
     parsedb_parser.add_argument("-k", "--keggorg", metavar='', type=str, default='-', help="A single KEGG Organism code. If provided, it takes precedence over --eggnog.")
     parsedb_parser.add_argument("--goodbefore", metavar='', type=str, default='-', help="Syntax is {pure_mid}-{rid1}-{rid2}. From top to bottom, build the universe until reaction {rid1}, transport {rid2} and metabolite {pure_mid} are reached.")
     parsedb_parser.add_argument("--onlyauthor", metavar='', type=str, default='-', help="Build the universe by parsing contents of the specified author ID only. Contents affected by --goodbefore are parsed anyway.")
     parsedb_parser.add_argument("--nofigs", action='store_true', help="Do not generate figures.")
     parsedb_parser.add_argument("-j", "--justparse", action='store_true', help="Just parse the database without performing extra activities (saves time during universe expansion).")
+    parsedb_parser.add_argument("-d", "--keepdisconn", action='store_true', help="Do not remove disconnected metabolites.")
     
     
     

gsrap/__init__.py

@@ -75,12 +75,14 @@ def main():
     parsedb_parser.add_argument("-z", "--initialize", metavar='', type=str, default='-', help="Initialize the universe on the provided medium. By default, the first medium in --media is used. Use 'none' to avoid initialization.")
     parsedb_parser.add_argument("--precursors", action='store_true', help="Verify biosynthesis of biomass precursors and show blocked ones.")
     parsedb_parser.add_argument("--biosynth", action='store_true', help="Check biosynthesis of all metabolites and detect dead-ends.")
+    parsedb_parser.add_argument("-t", "--taxon", metavar='', type=str, default='-', help="High-level taxon of interest. If provided, it must follow the syntax '{level}:{name}', where {level} is 'kingdom' or 'phylum'.")
     parsedb_parser.add_argument("-e", "--eggnog", nargs='+', metavar='', type=str, default='-', help="Path to the optional eggnog-mapper annotation table(s).")
     parsedb_parser.add_argument("-k", "--keggorg", metavar='', type=str, default='-', help="A single KEGG Organism code. If provided, it takes precedence over --eggnog.")
     parsedb_parser.add_argument("--goodbefore", metavar='', type=str, default='-', help="Syntax is {pure_mid}-{rid1}-{rid2}. From top to bottom, build the universe until reaction {rid1}, transport {rid2} and metabolite {pure_mid} are reached.")
     parsedb_parser.add_argument("--onlyauthor", metavar='', type=str, default='-', help="Build the universe by parsing contents of the specified author ID only. Contents affected by --goodbefore are parsed anyway.")
     parsedb_parser.add_argument("--nofigs", action='store_true', help="Do not generate figures.")
     parsedb_parser.add_argument("-j", "--justparse", action='store_true', help="Just parse the database without performing extra activities (saves time during universe expansion).")
+    parsedb_parser.add_argument("-d", "--keepdisconn", action='store_true', help="Do not remove disconnected metabolites.")
     
     
     

gsrap/commons/.ipynb_checkpoints/downloads-checkpoint.py

@@ -243,7 +243,99 @@ def format_expansion(logger, eggnog):
     
 
 
-
-
-
-
+def check_taxon(logger, taxon, idcollection_dict):
+    
+    
+    # verify presence of needed assets
+    if 'ko_to_taxa' not in idcollection_dict.keys():
+        logger.error(f"Asset 'ko_to_taxa' not found in 'gsrap.maps'. Please update 'gsrap.maps' with 'gsrap getmaps'.")
+        return 1
+    
+    
+    # extract level and name
+    try: level, name = taxon.split(':')
+    except: 
+        logger.error(f"Provided --taxon is not well formatted: '{taxon}'.")
+        return 1
+    
+    
+    # compute available levels and check
+    avail_levels = set(['kingdom', 'phylum'])            
+    if level not in avail_levels:
+        logger.error(f"Provided level is not acceptable: '{level}' (see --taxon). Acceptable levels are {avail_levels}.")
+        return 1
+    
+    
+    # compute available taxa at input level
+    avail_taxa_at_level = set()
+    ko_to_taxa = idcollection_dict['ko_to_taxa']
+    for ko in ko_to_taxa.keys():
+        for taxon_name in ko_to_taxa[ko][level]:
+            avail_taxa_at_level.add(taxon_name)
+    if name not in avail_taxa_at_level:
+        logger.error(f"Provided taxon name is not acceptable: '{name}' (see --taxon). Acceptable taxon names for level '{level}' are {avail_taxa_at_level}.")
+        return 1
+    
+    
+    """
+    sorted(list(df.query("kingdom == 'Bacteria'")['phylum'].unique()))
+    ['Acidobacteriota',
+     'Actinomycetota',
+     'Alphaproteobacteria',
+     'Aquificota',
+     'Armatimonadota',
+     'Atribacterota',
+     'Bacilli',
+     'Bacteria incertae sedis',
+     'Bacteroidota',
+     'Balneolota',
+     'Bdellovibrionota',
+     'Betaproteobacteria',
+     'Caldisericota',
+     'Calditrichota',
+     'Campylobacterota',
+     'Chlamydiota',
+     'Chlorobiota',
+     'Chloroflexota',
+     'Chrysiogenota',
+     'Cloacimonadota',
+     'Clostridia',
+     'Coprothermobacterota',
+     'Cyanobacteriota',
+     'Deferribacterota',
+     'Deinococcota',
+     'Deltaproteobacteria',
+     'Dictyoglomota',
+     'Elusimicrobiota',
+     'Enterobacteria',
+     'Fibrobacterota',
+     'Fidelibacterota',
+     'Fusobacteriota',
+     'Gemmatimonadota',
+     'Ignavibacteriota',
+     'Kiritimatiellota',
+     'Lentisphaerota',
+     'Melainabacteria',
+     'Mycoplasmatota',
+     'Myxococcota',
+     'Nitrospinota',
+     'Nitrospirota',
+     'Omnitrophota',
+     'Planctomycetota',
+     'Rhodothermota',
+     'Spirochaetota',
+     'Synergistota',
+     'Thermodesulfobacteriota',
+     'Thermodesulfobiota',
+     'Thermomicrobiota',
+     'Thermosulfidibacterota',
+     'Thermotogota',
+     'Verrucomicrobiota',
+     'Vulcanimicrobiota',
+     'other Bacillota',
+     'other Gammaproteobacteria',
+     'other Pseudomonadota',
+     'unclassified Bacteria']
+    """
+    
+    return 0

gsrap/commons/.ipynb_checkpoints/escherutils-checkpoint.py

@@ -1,3 +1,9 @@
+import warnings
+import logging
+
+
+import cobra
+
 
 
 def print_json_tree(data, level=0, max_level=2):
@@ -17,7 +23,7 @@ def print_json_tree(data, level=0, max_level=2):
             
             
 
-def count_undrawn_rids(logger, universe, lastmap):
+def count_undrawn_rids(logger, universe, lastmap, focus):
     
     
     rids = set([r.id for r in universe.reactions])
@@ -32,6 +38,71 @@ def count_undrawn_rids(logger, universe, lastmap):
     logger.debug(f"Last universal map version detected: '{filename}'.")
     if len(remainings) > 0:
         logger.warning(f"Our universal map is {len(remainings)} reactions behind. Please draw!")
+        if focus == '-':
+            logger.warning(f"Drawing is eased when using '--focus'...")
     else:
         logger.info(f"Our universal map is {len(remainings)} reactions behind. Thank you ♥")
+        
+        
+        
+def count_undrawn_rids_focus(logger, universe, lastmap, focus, outdir):
+    
+    
+    # get modeled reads for this --focus:
+    rids = set()
+    try: gr = universe.groups.get_by_id(focus)
+    except: 
+        logger.warning(f"Group '{focus}' not found!")
+        return 
+    for r in gr.members:
+        rids.add(r.id)
+
+
+    # get rids on Escher: 
+    drawn_rids = set()
+    for key, value in lastmap['json'][1]['reactions'].items():
+        drawn_rids.add(value['bigg_id'])
+
+        
+    # get remaining rids for this map:
+    remainings = rids - drawn_rids
+    remainings_krs = set()
+    for rid in remainings: 
+        r = universe.reactions.get_by_id(rid)
+        krs = r.annotation['kegg.reaction']
+        for kr in krs: 
+            remainings_krs.add(kr)
+    
+    
+    if len(remainings) > 0:
+        if focus != 'transport':
+            logger.warning(f"Focusing on '{focus}', our universal map is {len(remainings)} reactions behind: {' '.join(list(remainings_krs))}.")
+        else:
+            logger.warning(f"Focusing on '{focus}', our universal map is {len(remainings)} reactions behind.")  # usually no kegg codes for tranport reactions
+        
+        
+        # subset the universe to ease the drawing: 
+        universe_focus = universe.copy()
+        to_remove = [r for r in universe_focus.reactions if r.id not in rids]
+        
+        
+        # trick to avoid the WARNING "cobra/core/group.py:147: UserWarning: need to pass in a list" 
+        # triggered when trying to remove reactions that are included in groups. 
+        with warnings.catch_warnings():  # temporarily suppress warnings for this block
+            warnings.simplefilter("ignore")  # ignore all warnings
+            cobra_logger = logging.getLogger("cobra.util.solver")
+            old_level = cobra_logger.level
+            cobra_logger.setLevel(logging.ERROR)   
+           
+            universe_focus.remove_reactions(to_remove,remove_orphans=True)
+
+            # restore original behaviour: 
+            cobra_logger.setLevel(old_level)  
+            
+            
+        # save the subset for drawing in Escher!
+        logger.info(f"Writing '{outdir}/{focus}.json' to ease your drawing workflow...")
+        cobra.io.save_json_model(universe_focus, f'{outdir}/{focus}.json')
+    else:
+        logger.info(f"Focusing on '{focus}', our universal map is {len(remainings)} reactions behind. Thank you ♥")
         

gsrap/commons/.ipynb_checkpoints/excelhub-checkpoint.py

@@ -148,7 +148,7 @@ def write_excel_model(model, filepath, nofigs, memote_results_dict, df_E, df_B,
             else: df_T.append(row_dict)
             
     for g in model.genes:
-        row_dict = {'gid': g.id, 'involved_in': '; '.join([r.id for r in g.reactions])}
+        row_dict = {'gid': g.id, 'name': g.name, 'involved_in': '; '.join([r.id for r in g.reactions])}
         
         for db in g.annotation.keys():
             annots = g.annotation[db]
@@ -171,7 +171,7 @@ def write_excel_model(model, filepath, nofigs, memote_results_dict, df_E, df_B,
     df_R = df_R[df_R_first_cols + sorted([c for c in df_R.columns if c not in df_R_first_cols])]
     df_T = df_T[df_R_first_cols + sorted([c for c in df_T.columns if c not in df_R_first_cols])]
     df_A = df_A[df_R_first_cols + sorted([c for c in df_A.columns if c not in df_R_first_cols])]
-    df_G_first_cols = ['gid', 'involved_in']
+    df_G_first_cols = ['gid', 'name', 'involved_in']
     df_G = df_G[df_G_first_cols + sorted([c for c in df_G.columns if c not in df_G_first_cols])]
     
     

gsrap/commons/downloads.py

@@ -243,7 +243,99 @@ def format_expansion(logger, eggnog):
     
 
 
-
-
-
-
+def check_taxon(logger, taxon, idcollection_dict):
+    
+    
+    # verify presence of needed assets
+    if 'ko_to_taxa' not in idcollection_dict.keys():
+        logger.error(f"Asset 'ko_to_taxa' not found in 'gsrap.maps'. Please update 'gsrap.maps' with 'gsrap getmaps'.")
+        return 1
+    
+    
+    # extract level and name
+    try: level, name = taxon.split(':')
+    except: 
+        logger.error(f"Provided --taxon is not well formatted: '{taxon}'.")
+        return 1
+    
+    
+    # compute available levels and check
+    avail_levels = set(['kingdom', 'phylum'])            
+    if level not in avail_levels:
+        logger.error(f"Provided level is not acceptable: '{level}' (see --taxon). Acceptable levels are {avail_levels}.")
+        return 1
+    
+    
+    # compute available taxa at input level
+    avail_taxa_at_level = set()
+    ko_to_taxa = idcollection_dict['ko_to_taxa']
+    for ko in ko_to_taxa.keys():
+        for taxon_name in ko_to_taxa[ko][level]:
+            avail_taxa_at_level.add(taxon_name)
+    if name not in avail_taxa_at_level:
+        logger.error(f"Provided taxon name is not acceptable: '{name}' (see --taxon). Acceptable taxon names for level '{level}' are {avail_taxa_at_level}.")
+        return 1
+    
+    
+    """
+    sorted(list(df.query("kingdom == 'Bacteria'")['phylum'].unique()))
+    ['Acidobacteriota',
+     'Actinomycetota',
+     'Alphaproteobacteria',
+     'Aquificota',
+     'Armatimonadota',
+     'Atribacterota',
+     'Bacilli',
+     'Bacteria incertae sedis',
+     'Bacteroidota',
+     'Balneolota',
+     'Bdellovibrionota',
+     'Betaproteobacteria',
+     'Caldisericota',
+     'Calditrichota',
+     'Campylobacterota',
+     'Chlamydiota',
+     'Chlorobiota',
+     'Chloroflexota',
+     'Chrysiogenota',
+     'Cloacimonadota',
+     'Clostridia',
+     'Coprothermobacterota',
+     'Cyanobacteriota',
+     'Deferribacterota',
+     'Deinococcota',
+     'Deltaproteobacteria',
+     'Dictyoglomota',
+     'Elusimicrobiota',
+     'Enterobacteria',
+     'Fibrobacterota',
+     'Fidelibacterota',
+     'Fusobacteriota',
+     'Gemmatimonadota',
+     'Ignavibacteriota',
+     'Kiritimatiellota',
+     'Lentisphaerota',
+     'Melainabacteria',
+     'Mycoplasmatota',
+     'Myxococcota',
+     'Nitrospinota',
+     'Nitrospirota',
+     'Omnitrophota',
+     'Planctomycetota',
+     'Rhodothermota',
+     'Spirochaetota',
+     'Synergistota',
+     'Thermodesulfobacteriota',
+     'Thermodesulfobiota',
+     'Thermomicrobiota',
+     'Thermosulfidibacterota',
+     'Thermotogota',
+     'Verrucomicrobiota',
+     'Vulcanimicrobiota',
+     'other Bacillota',
+     'other Gammaproteobacteria',
+     'other Pseudomonadota',
+     'unclassified Bacteria']
+    """
+    
+    return 0

gsrap/commons/escherutils.py

@@ -1,3 +1,9 @@
+import warnings
+import logging
+
+
+import cobra
+
 
 
 def print_json_tree(data, level=0, max_level=2):
@@ -17,7 +23,7 @@ def print_json_tree(data, level=0, max_level=2):
             
             
 
-def count_undrawn_rids(logger, universe, lastmap):
+def count_undrawn_rids(logger, universe, lastmap, focus):
     
     
     rids = set([r.id for r in universe.reactions])
@@ -32,6 +38,71 @@ def count_undrawn_rids(logger, universe, lastmap):
     logger.debug(f"Last universal map version detected: '{filename}'.")
     if len(remainings) > 0:
         logger.warning(f"Our universal map is {len(remainings)} reactions behind. Please draw!")
+        if focus == '-':
+            logger.warning(f"Drawing is eased when using '--focus'...")
     else:
         logger.info(f"Our universal map is {len(remainings)} reactions behind. Thank you ♥")
+        
+        
+        
+def count_undrawn_rids_focus(logger, universe, lastmap, focus, outdir):
+    
+    
+    # get modeled reads for this --focus:
+    rids = set()
+    try: gr = universe.groups.get_by_id(focus)
+    except: 
+        logger.warning(f"Group '{focus}' not found!")
+        return 
+    for r in gr.members:
+        rids.add(r.id)
+
+
+    # get rids on Escher: 
+    drawn_rids = set()
+    for key, value in lastmap['json'][1]['reactions'].items():
+        drawn_rids.add(value['bigg_id'])
+
+        
+    # get remaining rids for this map:
+    remainings = rids - drawn_rids
+    remainings_krs = set()
+    for rid in remainings: 
+        r = universe.reactions.get_by_id(rid)
+        krs = r.annotation['kegg.reaction']
+        for kr in krs: 
+            remainings_krs.add(kr)
+    
+    
+    if len(remainings) > 0:
+        if focus != 'transport':
+            logger.warning(f"Focusing on '{focus}', our universal map is {len(remainings)} reactions behind: {' '.join(list(remainings_krs))}.")
+        else:
+            logger.warning(f"Focusing on '{focus}', our universal map is {len(remainings)} reactions behind.")  # usually no kegg codes for tranport reactions
+        
+        
+        # subset the universe to ease the drawing: 
+        universe_focus = universe.copy()
+        to_remove = [r for r in universe_focus.reactions if r.id not in rids]
+        
+        
+        # trick to avoid the WARNING "cobra/core/group.py:147: UserWarning: need to pass in a list" 
+        # triggered when trying to remove reactions that are included in groups. 
+        with warnings.catch_warnings():  # temporarily suppress warnings for this block
+            warnings.simplefilter("ignore")  # ignore all warnings
+            cobra_logger = logging.getLogger("cobra.util.solver")
+            old_level = cobra_logger.level
+            cobra_logger.setLevel(logging.ERROR)   
+           
+            universe_focus.remove_reactions(to_remove,remove_orphans=True)
+
+            # restore original behaviour: 
+            cobra_logger.setLevel(old_level)  
+            
+            
+        # save the subset for drawing in Escher!
+        logger.info(f"Writing '{outdir}/{focus}.json' to ease your drawing workflow...")
+        cobra.io.save_json_model(universe_focus, f'{outdir}/{focus}.json')
+    else:
+        logger.info(f"Focusing on '{focus}', our universal map is {len(remainings)} reactions behind. Thank you ♥")
         

gsrap/commons/excelhub.py

@@ -148,7 +148,7 @@ def write_excel_model(model, filepath, nofigs, memote_results_dict, df_E, df_B,
             else: df_T.append(row_dict)
             
     for g in model.genes:
-        row_dict = {'gid': g.id, 'involved_in': '; '.join([r.id for r in g.reactions])}
+        row_dict = {'gid': g.id, 'name': g.name, 'involved_in': '; '.join([r.id for r in g.reactions])}
         
         for db in g.annotation.keys():
             annots = g.annotation[db]
@@ -171,7 +171,7 @@ def write_excel_model(model, filepath, nofigs, memote_results_dict, df_E, df_B,
     df_R = df_R[df_R_first_cols + sorted([c for c in df_R.columns if c not in df_R_first_cols])]
     df_T = df_T[df_R_first_cols + sorted([c for c in df_T.columns if c not in df_R_first_cols])]
     df_A = df_A[df_R_first_cols + sorted([c for c in df_A.columns if c not in df_R_first_cols])]
-    df_G_first_cols = ['gid', 'involved_in']
+    df_G_first_cols = ['gid', 'name', 'involved_in']
     df_G = df_G[df_G_first_cols + sorted([c for c in df_G.columns if c not in df_G_first_cols])]
     
     

gsrap/getmaps/.ipynb_checkpoints/getmaps-checkpoint.py

@@ -4,6 +4,7 @@ import pickle
 
 
 from .kdown import download_raw_txtfiles
+from .kdown import create_dict_keggorg
 from .kdown import create_dict_ko
 from .kdown import create_dict_c
 from .kdown import create_dict_r
@@ -20,13 +21,19 @@ def do_kdown(logger, outdir, usecache, keeptmp):
     logger.info(f"Respectfully retrieving metabolic information from KEGG. Raw data are being saved into '{outdir}/kdown/'. Be patient, could take a couple of days...")
     os.makedirs(f'{outdir}/kdown/', exist_ok=True)
     
+    
     response = download_raw_txtfiles(logger, outdir, usecache)
     if type(response) == int: return 1
     else: RELEASE_kegg = response
     
+        
     
     logger.info("Parsing downloaded KEGG information...")
-
+    
+    response = create_dict_keggorg(logger, outdir)
+    if type(response) == int: return 1
+    else: dict_keggorg = response
+    
     response = create_dict_ko(logger, outdir)
     if type(response) == int: return 1
     else: dict_ko = response
@@ -49,7 +56,7 @@ def do_kdown(logger, outdir, usecache, keeptmp):
     
     
     # create 'idcollection_dict' and 'summary_dict' dictionaries
-    idcollection_dict = create_idcollection_dict(dict_ko, dict_c, dict_r, dict_map, dict_md)
+    idcollection_dict = create_idcollection_dict(dict_keggorg, dict_ko, dict_c, dict_r, dict_map, dict_md)
     summary_dict = create_summary_dict(dict_c, dict_r, dict_map, dict_md)
     
         
@@ -57,7 +64,6 @@ def do_kdown(logger, outdir, usecache, keeptmp):
 
 
 
-
 def main(args, logger):
     
     
@@ -67,7 +73,7 @@ def main(args, logger):
     os.makedirs(f'{args.outdir}/', exist_ok=True)
     
     
-    # KEGG
+    # KEGG download
     response = do_kdown(logger, args.outdir, args.usecache, args.keeptmp)
     if type(response) == int: return 1
     else: RELEASE_kegg, idcollection_dict, summary_dict = response[0], response[1], response[2]
@@ -76,7 +82,9 @@ def main(args, logger):
     # create 'gsrap.maps':
     with open(f'{args.outdir}/gsrap.maps', 'wb') as wb_handler:
         pickle.dump({
-            'RELEASE_kegg': RELEASE_kegg, 'idcollection_dict': idcollection_dict, 'summary_dict': summary_dict,
+            'RELEASE_kegg': RELEASE_kegg, 
+            'idcollection_dict': idcollection_dict, 
+            'summary_dict': summary_dict, 
         }, wb_handler)
     logger.info(f"'{args.outdir}/gsrap.maps' created!")
     
@@ -87,4 +95,5 @@ def main(args, logger):
         logger.info(f"Temporary raw files deleted!")
     
     
+    
     return 0

gsrap/getmaps/.ipynb_checkpoints/kdown-checkpoint.py

@@ -34,6 +34,7 @@ def download_raw_txtfiles(logger, outdir, usecache):
         'orthology',
         'module', 
         'pathway',
+        'organism',
     ]
     for db in databases:
         time.sleep(0.5)
@@ -45,8 +46,9 @@ def download_raw_txtfiles(logger, outdir, usecache):
 
     # mix the items to download to be respectful/compliant
     items_to_download = []
-
     for db in databases:
+        if db == 'organism': 
+            continue   # here we just need the list
         with open(f"{outdir}/kdown/{db}.txt", 'r') as file:
             res_string = file.read()
             rows = res_string.split('\n')
@@ -54,7 +56,6 @@ def download_raw_txtfiles(logger, outdir, usecache):
                 item_id = row.split('\t', 1)[0]
                 if item_id == '': continue
                 items_to_download.append({'db': db, 'id': item_id})
-
     random.shuffle(items_to_download)
 
 
@@ -79,6 +80,51 @@ def download_raw_txtfiles(logger, outdir, usecache):
 
 
 
+def create_dict_keggorg(logger, outdir):
+    
+    organisms_raw = open(f'{outdir}/kdown/organism.txt', 'r').read()
+
+    # create a dataframe listing all organisms in KEGG;
+    # columns are [tnumber, name, domain, kingdom, phylum, classification]
+    df = []  # list fo dicts
+    for line in organisms_raw.strip().split("\n"):
+        fields = line.split("\t")
+        if len(fields) == 4:
+            tnumber, keggorg, name, classification = fields
+            levels = classification.split(";")
+            domain = levels[0]
+            kingdom = levels[1]
+            phylum = levels[2]
+            df.append({
+                'tnumber':tnumber, 
+                'keggorg': keggorg, 
+                'name': name, 
+                'domain': domain, 
+                'kingdom': kingdom,
+                'phylum': phylum,
+                'classification': classification
+            })
+        else: 
+            # never verified during tests!
+            logger.warning(f'Strange number of fields found in this line of "organism.txt": """{line}""".')
+    df = pnd.DataFrame.from_records(df)
+    df = df.set_index('keggorg', drop=True, verify_integrity=True)
+    
+    
+    # convert dataframe to dict
+    dict_keggorg = {}
+    for keggorg, row in df.iterrows(): 
+        dict_keggorg[keggorg] = {
+            'kingdom': row['kingdom'],
+            'phylum': row['phylum'],
+            #'name': row['name'],   # not strictly needed. Commented to save disk space.
+        }
+        
+    if logger != None: logger.info(f'Number of unique items (org): {len(dict_keggorg.keys())}.')
+    return dict_keggorg
+
+
+
 def create_dict_ko(logger, outdir):
     
     dict_ko = {}         # main output
@@ -98,6 +144,7 @@ def create_dict_ko(logger, outdir):
                 'ecs': set(),
                 'cogs': set(),
                 'gos': set(),
+                'keggorgs': set(),
             }   
         else:
             logger.error(f"{ko_id} already included!")
@@ -175,7 +222,13 @@ def create_dict_ko(logger, outdir):
                         gos = content[len('GO: '):].strip().split(' ')
                         for go in gos:
                             dict_ko[ko_id]['gos'].add(go)
-                        
+                            
+                            
+                # parse the organism-specific genes
+                if curr_header == 'GENES       ':
+                    keggorg = content.split(': ',1)[0]
+                    dict_ko[ko_id]['keggorgs'].add(keggorg.lower()) # organism.txt has IDs in lowercase
+                    
                         
                 # parse the reactions
                 if curr_header == 'REACTION    ':
@@ -547,7 +600,7 @@ def create_dict_md(logger, outdir):
 
 
 
-def create_idcollection_dict(dict_ko, dict_c, dict_r, dict_map, dict_md):
+def create_idcollection_dict(dict_keggorg, dict_ko, dict_c, dict_r, dict_map, dict_md):
 
     idcollection_dict = {}
     
@@ -620,6 +673,24 @@ def create_idcollection_dict(dict_ko, dict_c, dict_r, dict_map, dict_md):
         for go in dict_ko[ko_id]['gos']:
             idcollection_dict['ko_to_gos'][ko_id].add(go)
             
+    
+    # creation of 'ko_to_keggorgs' skipped as it takes too much disk space. Replaced with 'ko_to_taxa'. 
+    idcollection_dict['ko_to_taxa'] = {}
+    missing_keggorgs = set()
+    for ko_id in dict_ko.keys():
+        idcollection_dict['ko_to_taxa'][ko_id] = {'kingdom': set(), 'phylum': set()}
+        for keggorg in dict_ko[ko_id]['keggorgs']:
+            try: 
+                kingdom = dict_keggorg[keggorg]['kingdom']
+                phylum = dict_keggorg[keggorg]['phylum']
+            except: 
+                if keggorg not in missing_keggorgs:
+                    missing_keggorgs.add(keggorg)
+                    #print(f"Organism '{keggorg}' appears in 'orthology/' but not in 'organism.txt'.")
+                continue
+            idcollection_dict['ko_to_taxa'][ko_id]['kingdom'].add(kingdom)
+            idcollection_dict['ko_to_taxa'][ko_id]['phylum'].add(phylum)
+
             
     idcollection_dict['map_to_name'] = {}
     for map_id in dict_map.keys():