Explore Reactome

Exploring Reactome from a network perspective

Loading the node and edge info

from lipinet.parse_reactome import parse_reactome_data 

reactome_results = parse_reactome_data(verbose=True, use_cache=True)
df_reactome_nodes = reactome_results['df_nodes']
df_reactome_edges = reactome_results['df_edges']

↪ loading Reactome (processed) from cache: reactome_human_nb

df_reactome_nodes

	node_id	layer	source_db_identifier	reactome_pe_name	species	pe_name	pe_location	human	url	event_name_pathway_or_reaction	evidence_code	name
0	10033	reactome_chebi	NaN	NaN	NaN	<NA>	<NA>	NaN	NaN	NaN	NaN	NaN
1	10036	reactome_chebi	NaN	NaN	NaN	<NA>	<NA>	NaN	NaN	NaN	NaN	NaN
2	10055	reactome_chebi	NaN	NaN	NaN	<NA>	<NA>	NaN	NaN	NaN	NaN	NaN
3	10093	reactome_chebi	NaN	NaN	NaN	<NA>	<NA>	NaN	NaN	NaN	NaN	NaN
4	10100	reactome_chebi	NaN	NaN	NaN	<NA>	<NA>	NaN	NaN	NaN	NaN	NaN
...	...	...	...	...	...	...	...	...	...	...	...	...
169357	autophagosome membrane	reactome_physicalent_loc	NaN	NaN	NaN	<NA>	<NA>	NaN	NaN	NaN	NaN	NaN
169358	lamellar body	reactome_physicalent_loc	NaN	NaN	NaN	<NA>	<NA>	NaN	NaN	NaN	NaN	NaN
169359	lamellar body membrane	reactome_physicalent_loc	NaN	NaN	NaN	<NA>	<NA>	NaN	NaN	NaN	NaN	NaN
169360	clathrin-sculpted gamma-aminobutyric acid tran...	reactome_physicalent_loc	NaN	NaN	NaN	<NA>	<NA>	NaN	NaN	NaN	NaN	NaN
169361	endosome	reactome_physicalent_loc	NaN	NaN	NaN	<NA>	<NA>	NaN	NaN	NaN	NaN	NaN

55191 rows × 12 columns

df_reactome_edges

	source_id	target_id	source_layer	target_layer	source_db_identifier	reactome_pe_name	url	event_name_pathway_or_reaction	evidence_code	species	human	pe_name	pe_location	name	interlayer
0	10033	R-ALL-9014945	reactome_chebi	reactome_physicalent	NaN	NaN	NaN	NaN	NaN	NaN	NaN	<NA>	<NA>	NaN	NaN
1	10036	R-ALL-5696412	reactome_chebi	reactome_physicalent	NaN	NaN	NaN	NaN	NaN	NaN	NaN	<NA>	<NA>	NaN	NaN
2	10055	R-ALL-9611688	reactome_chebi	reactome_physicalent	NaN	NaN	NaN	NaN	NaN	NaN	NaN	<NA>	<NA>	NaN	NaN
3	10093	R-ALL-9648287	reactome_chebi	reactome_physicalent	NaN	NaN	NaN	NaN	NaN	NaN	NaN	<NA>	<NA>	NaN	NaN
4	10093	R-ALL-3296452	reactome_chebi	reactome_physicalent	NaN	NaN	NaN	NaN	NaN	NaN	NaN	<NA>	<NA>	NaN	NaN
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
697167	R-XTR-9958790	R-XTR-427652	reactome_pathway_ontology	reactome_pathway_ontology	NaN	NaN	NaN	NaN	NaN	NaN	NaN	<NA>	<NA>	NaN	False
697168	R-XTR-9958790	R-XTR-433137	reactome_pathway_ontology	reactome_pathway_ontology	NaN	NaN	NaN	NaN	NaN	NaN	NaN	<NA>	<NA>	NaN	False
697169	R-XTR-9958863	R-XTR-352230	reactome_pathway_ontology	reactome_pathway_ontology	NaN	NaN	NaN	NaN	NaN	NaN	NaN	<NA>	<NA>	NaN	False
697170	R-XTR-9958863	R-XTR-428559	reactome_pathway_ontology	reactome_pathway_ontology	NaN	NaN	NaN	NaN	NaN	NaN	NaN	<NA>	<NA>	NaN	False
697171	R-XTR-9959399	R-XTR-427975	reactome_pathway_ontology	reactome_pathway_ontology	NaN	NaN	NaN	NaN	NaN	NaN	NaN	<NA>	<NA>	NaN	False

164767 rows × 15 columns

from graph_tool.all import Graph, GraphView, graph_draw
import graph_tool as gt

from onionnet import OnionNet
import onionnet.visualisation

import pandas as pd

import onionnet.exporter 

Building the network with OnionNet

onion = OnionNet()

onion.grow_onion(df_nodes=df_reactome_nodes,
           df_edges=df_reactome_edges,
           node_prop_cols=df_reactome_nodes.columns.to_list(),
           edge_prop_cols=df_reactome_edges.columns.to_list(),
           drop_na=True,
           drop_duplicates=True)

Nodes: in=55191, dropped_na=0, deduped=1017 → final=54174
Edges: in=164767, dropped_invalid=3018, deduped=8285 → final=133171

Note that we have some dedupes, would need to carefully consider if this has led to a loss of information or corruption.

onion.core.graph

<Graph object, directed, with 54174 vertices and 133171 edges, 14 internal vertex properties, 15 internal edge properties, at 0x3313b26c0>

list(onion.core.graph.vp)

['layer_hash',
 'node_id_hash',
 'node_id',
 'layer',
 'source_db_identifier',
 'reactome_pe_name',
 'species',
 'pe_name',
 'pe_location',
 'human',
 'url',
 'event_name_pathway_or_reaction',
 'evidence_code',
 'name']

df_reactome_nodes.info()

<class 'pandas.core.frame.DataFrame'>
Index: 55191 entries, 0 to 169361
Data columns (total 12 columns):
 #   Column                          Non-Null Count  Dtype  
---  ------                          --------------  -----  
 0   node_id                         55191 non-null  object 
 1   layer                           55191 non-null  object 
 2   source_db_identifier            6277 non-null   float64
 3   reactome_pe_name                6277 non-null   object 
 4   species                         42032 non-null  object 
 5   pe_name                         12199 non-null  string 
 6   pe_location                     12199 non-null  string 
 7   human                           18875 non-null  object 
 8   url                             12598 non-null  object 
 9   event_name_pathway_or_reaction  12598 non-null  object 
 10  evidence_code                   12598 non-null  object 
 11  name                            26299 non-null  object 
dtypes: float64(1), object(9), string(2)
memory usage: 5.5+ MB

Decode property labels that were encoded

onion.decode_property_labels_bulk(df=df_reactome_nodes, encoded_prop_type='v')

V property 'node_id_decoded' created successfully.
V property 'layer_decoded' created successfully.
source_db_identifier prop left as is, no decoding needed (not an object type)
V property 'reactome_pe_name_decoded' created successfully.
V property 'species_decoded' created successfully.
pe_name prop left as is, no decoding needed (not an object type)
pe_location prop left as is, no decoding needed (not an object type)
V property 'human_decoded' created successfully.
V property 'url_decoded' created successfully.
V property 'event_name_pathway_or_reaction_decoded' created successfully.
V property 'evidence_code_decoded' created successfully.
V property 'name_decoded' created successfully.

onion.decode_property_labels_bulk(df=df_reactome_edges, encoded_prop_type='e')

E property 'source_id_decoded' created successfully.
E property 'target_id_decoded' created successfully.
E property 'ec_level_decoded' created successfully.
E property 'source_layer_decoded' created successfully.
E property 'target_layer_decoded' created successfully.
interlayer prop left as is, no decoding needed (not an object type)
E property 'edge_type_decoded' created successfully.

df_reactome_edges.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 108439 entries, 0 to 108438
Data columns (total 7 columns):
 #   Column        Non-Null Count   Dtype 
---  ------        --------------   ----- 
 0   source_id     108439 non-null  object
 1   target_id     98225 non-null   object
 2   ec_level      6482 non-null    object
 3   source_layer  108439 non-null  object
 4   target_layer  108439 non-null  object
 5   interlayer    108439 non-null  bool  
 6   edge_type     108439 non-null  object
dtypes: bool(1), object(6)
memory usage: 5.1+ MB

onion.core.layer_code_to_name

{0: 'rhea_reactionid', 1: 'rhea_chebiid', 2: 'rhea_ec'}

Inspecting LipiNet metagraph for Reactome

from onionnet.analytics import layer_stats, plot_layer_metagraph

nodes_by_layer, edges_by_pair = layer_stats(
    df_nodes=df_reactome_nodes,
    df_edges=df_reactome_edges,
    print_tables=True
)

Node counts by layer:

	count
layer
reactome_pathway_ontology	23157
reactome_reactions	9456
reactome_physicalent	6277
reactome_physicalent_nameloc	5922
reactome_physicalent_name	4084
reactome_pathway	3142
reactome_chebi	3068
reactome_physicalent_loc	85

Interlayer edge count: 0
Edge counts by (source_layer, target_layer):

		edges
source_layer	target_layer
reactome_physicalent	reactome_pathway	57728
	reactome_reactions	35704
reactome_pathway_ontology	reactome_pathway_ontology	23259
reactome_pathway	reactome_ontology_pathways	20860
reactome_chebi	reactome_physicalent	6353
reactome_physicalent_nameloc	reactome_physicalent	6001
reactome_physicalent_loc	reactome_physicalent_nameloc	5922
reactome_physicalent_name	reactome_physicalent_nameloc	5922

mg, mg_pos = plot_layer_metagraph(
    edges_by_pair,
    nodes_by_layer=nodes_by_layer,
    node_size_range=(20, 30),
    node_text_size_range=(12, 16),
    edge_width_range=(6, 12),
    node_scaler="log",
    edge_scaler="log",
    show_labels=True,
    output_size=(1000, 1000),
    node_text_position=1,
    return_graph=True,
    # pad_label_string=True,
    # vertex_font='consolas',
    # family_extractor=my_family,
    # family_colors={"sl": (0.2,0.6,0.9,0.9), "rhea": (0.9,0.4,0.1,0.9)}
)

More TODO - notebook in progress