AnnNet Showcase¶

This notebook is a compact tour of the current public API, with particular emphasis on the edge model.

The central fact to keep in mind is this: add_edge(...) is not just for binary edges. It is the unified structural constructor for the graph.

In [ ]:

from pathlib import Path
import sys

repo_root = Path.cwd()
if not (repo_root / 'annnet').exists():
    for parent in repo_root.parents:
        if (parent / 'annnet').exists():
            repo_root = parent
            break
if str(repo_root) not in sys.path:
    sys.path.insert(0, str(repo_root))

import annnet as an

from pathlib import Path import sys repo_root = Path.cwd() if not (repo_root / 'annnet').exists(): for parent in repo_root.parents: if (parent / 'annnet').exists(): repo_root = parent break if str(repo_root) not in sys.path: sys.path.insert(0, str(repo_root)) import annnet as an

In [ ]:

G = an.AnnNet(directed=True)
G.add_vertices_bulk(
    [
        ('A', {'kind': 'gene'}),
        ('B', {'kind': 'gene'}),
        ('C', {'kind': 'gene'}),
        ('D', {'kind': 'metabolite'}),
        ('E', {'kind': 'complex'}),
    ]
)

G = an.AnnNet(directed=True) G.add_vertices_bulk( [ ('A', {'kind': 'gene'}), ('B', {'kind': 'gene'}), ('C', {'kind': 'gene'}), ('D', {'kind': 'metabolite'}), ('E', {'kind': 'complex'}), ] )

Binary, hyper, stoichiometric, and edge-entity cases¶

All of these go through add_edge(...), but the endpoint shapes differ.

In [ ]:

# binary
G.add_edge('A', 'B', edge_id='e_bin', directed=True, relation='activates')

# undirected hyperedge
G.add_edge(src=['A', 'B', 'C'], edge_id='e_hyper', directed=False, relation='complex')

# directed hyperedge
G.add_edge(src=['A', 'B'], tgt=['D'], edge_id='e_dir_hyper', directed=True, relation='reaction')

# explicit stoichiometric incidence
G.add_edge(src={'A': 2.0, 'B': 1.0}, tgt={'D': 1.0}, edge_id='e_stoich', directed=True)

# edge-entity placeholder and edge->vertex relation
G.add_edge(edge_id='meta_e', as_entity=True, role='edge entity placeholder')
G.add_edge('meta_e', 'E', edge_id='e_meta_link', directed=True, edge_type='vertex_edge')

G.edges_view()

# binary G.add_edge('A', 'B', edge_id='e_bin', directed=True, relation='activates') # undirected hyperedge G.add_edge(src=['A', 'B', 'C'], edge_id='e_hyper', directed=False, relation='complex') # directed hyperedge G.add_edge(src=['A', 'B'], tgt=['D'], edge_id='e_dir_hyper', directed=True, relation='reaction') # explicit stoichiometric incidence G.add_edge(src={'A': 2.0, 'B': 1.0}, tgt={'D': 1.0}, edge_id='e_stoich', directed=True) # edge-entity placeholder and edge->vertex relation G.add_edge(edge_id='meta_e', as_entity=True, role='edge entity placeholder') G.add_edge('meta_e', 'E', edge_id='e_meta_link', directed=True, edge_type='vertex_edge') G.edges_view()

Managers and caches¶

The manager namespaces are now the primary public route for specialized behavior.

In [ ]:

print('idx stats:', G.idx.stats())
print('cache info:', G.cache.info())
print('view edge count:', G.view(edges=['e_bin', 'e_hyper']).edge_count)

print('idx stats:', G.idx.stats()) print('cache info:', G.cache.info()) print('view edge count:', G.view(edges=['e_bin', 'e_hyper']).edge_count)