set_enrichment.py - cMonkey set_enrichment scoring.
 
This file is part of cMonkey Python. Please see README and LICENSE for
more information and licensing details.

Modules
		cmonkey.datamatrix json logging math multiprocessing numpy os cmonkey.scoring cmonkey.util

Classes
		cmonkey.scoring.ScoringFunctionBase ScoringFunction CutoffEnrichmentSet DiscreteEnrichmentSet SetType   class CutoffEnrichmentSet     Enrichment set representation constructed with a cutoff     Methods defined here: __init__(self, cutoff, elems) instance creation __repr__(self) genes(self) returns all genes genes_above_cutoff(self) returns the genes that have a weight above the cutoff   class DiscreteEnrichmentSet       Methods defined here: __init__(self, genes) instance creation __repr__(self) genes(self) genes_above_cutoff(self) returns the genes that have a weight above the cutoff   class ScoringFunction(cmonkey.scoring.ScoringFunctionBase)     Set enrichment scoring function     Methods defined here: __init__(self, organism, membership, ratios, config_params=None) Create scoring function instance bonferroni_cutoff(self) Bonferroni cutoff value do_compute(self, iteration_result, ref_matrix) compute method Note: will return None if not computed yet and the result of a previous scoring if the function is not supposed to actually run in this iteration run_logs(self) return the run logs 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 compute(self, iteration_result, reference_matrix=None) general compute method, iteration_result is a dictionary that contains the results generated by the scoring functions in the current computation. the reference_matrix is actually a hack that allows the scoring function to normalize its scores to the range of a reference score matrix. In the normal case, those would be the gene expression row scores compute_force(self, iteration_result, reference_matrix=None) enforce computation, regardless of the iteration function 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)   class SetType     Set type representation. This is just a grouping from name to a number of sets and providing access to all contained genes     Methods defined here: __init__(self, name, sets) instance creation __repr__(self) string representation genes(self) All genes contained in the sets

Functions
		compute_cluster_score(args) Computes the cluster score for a given set type process_sets(input_sets, thesaurus) Reusable function that maps a dictionary {name: [genes]} to a set of DiscretenEnrichmentSet objects read_set_types(config_params, thesaurus) Reads sets from a JSON file. We also ensure that genes are stored in canonical form in the set, so that set operations based on gene names will succeed read_sets_csv(infile, thesaurus, sep=',') Reads sets from a CSV file We support 2-column and 3 column formats:   2 column files have the format   <set name><separator><gene>   3 column files have the format   <set name><separator><gene><weight>

Data
		CANONICAL_ROWNAMES = None CANONICAL_ROW_INDEXES = None SET_MATRIX = None SET_MEMBERSHIP = None SET_SET_TYPE = None SET_SYNONYMS = None