| | |
- Network
- cmonkey.scoring.ScoringFunctionBase
-
- ScoringFunction
class Network |
| |
class to represent a network graph.
The graph is considered undirected
For efficiency reasons, edges is a list of [source, target, weight]
|
| |
Methods defined here:
- __init__(self, name, edges, weight, dummy)
- creates a network from a list of edges
- __repr__(self)
- edges_with_node(self, node)
- returns the edges where node is a node of
- normalize_scores_to(self, score)
- normalizes all edge scores so that they sum up to
the specified score
- num_edges(self)
- returns the number of edges in this graph
- total_score(self)
- returns the sum of edge scores
- validate(self, synonyms, genes)
Class methods defined here:
- create(cls, name, edges, weight, organism=None, ratios=None, check_size=True) from __builtin__.classobj
- standard Factory method
|
class ScoringFunction(cmonkey.scoring.ScoringFunctionBase) |
| |
Network scoring function. Note that even though there are several
networks, scoring can't be generalized with the default ScoringCombiner,
since the scores are computed through weighted addition rather than
quantile normalization
|
| |
Methods defined here:
- __init__(self, organism, membership, ratios, config_params)
- Create scoring function instance
- compute(self, iteration_result, ref_matrix=None)
- overridden compute for storing additional information
- compute_force(self, iteration_result, ref_matrix=None)
- overridden compute for storing additional information
- do_compute(self, iteration_result, ref_matrix=None)
- compute method, iteration is the 0-based iteration number
- initialize(self, args)
- process additional parameters
- networks(self)
- networks are cached
- run_logs(self)
Methods inherited from cmonkey.scoring.ScoringFunctionBase:
- check_requirements(self)
- Give the scoring module an opportunity to check whether the
requirements to run are all met
- gene_names(self)
- returns the gene names
- last_cached(self)
- num_clusters(self)
- returns the number of clusters
- pickle_path(self)
- returns the function-specific pickle-path
- rows_for_cluster(self, cluster)
- returns the rows for the specified cluster
- run_in_iteration(self, i)
- scaling(self, iteration)
- returns the quantile normalization scaling for the specified iteration
- set_score_means(self, iteration_result, matrix)
| |