EM.py
 
This file contains class definitions for 5 classes useful for using EM for
Motif Discovery
 
 
class EM:                 Implementation of an EM
 
class MarkovBackground:   Implementation of a n-th order Markov Background model
                          (Pre-compute models using Background.py)
 
class Probe:              A probe (string with additional baggage)
 
class MotifCandidate:     Essentially a "super" Motif, that can compute its MAP score
                          probably should have been subclassed.
 
Copyright (2005) Whitehead Institute for Biomedical Research (except as noted below)
All Rights Reserved
 
Author: David Benjamin Gordon

Modules
		TAMO.seq.Fasta TAMO.MD.MDsupport TAMO.MotifTools TAMO math os re sys time

Classes
		EM MarkovBackground Zeroth MotifCandidate Probe   class EM       Methods defined here: EM_C(self, Model, probes, store_Zs='') EM_Cstart(self) __init__(self, seed_seqs, all_seqs, width=6, verbose='') all_Wmers(self, N, seq) best_by_Z(self, Wmers_by_seq) calcmask(self, width) compute_joint(self, model, Wmers_by_seq) report(self) seed_models(self)   class MarkovBackground       Methods defined here: ___getitem__(self, key) __init__(self, species='YEAST', seqs='') compute_conditional(self) freq_from_fasta(self, fastafile) freq_from_file(self) freq_from_seqs(self, seqs) freq_from_seqs_old(self, seqs) logbackground(self, seq) permute(self, letters, depth, seqs=[''], curdepth=0) study_seqs(self, seqs) zeroth(self)   class MotifCandidate     A candidate for Motif consists of the following:    A set of segments "wmers"    A pssm computed from these segments    A score (probably the MAP score described in the literature)    General propeties:      # segments      Width      A candidate must also be able to:      Evaluate its own score      Modify itself:        Add wmer        Remove wmer     Methods defined here: MAPpurge(self, verbose='') MAPscan(self, nmers) __init__(self, wmers='') __repr__(self) check_and_update(self, wmer, count, verbose='') computeMAP(self, in_wmers='') computeMAP_memo(self, wmer, count) find_wmers(self, seqs) has_wmer(self, wmer) logPbackground(self, wmer) purge(self, verbose='')   class Probe     Probe object: Extension of string class containing extra information, including logP and other information     Methods defined here: __cmp__(self, other) __commented_del__(self) __getitem__(self, n) __getslice__(self, i, j) __hash__(self) __init__(self, seq='') __len__(self) __repr__(self) compute_logP(self) translate(self, table)   class Zeroth(MarkovBackground)       Methods defined here: __init__(self, bgD={'A': 0.25, 'C': 0.25, 'G': 0.25, 'T': 0.25}) Methods inherited from MarkovBackground: ___getitem__(self, key) compute_conditional(self) freq_from_fasta(self, fastafile) freq_from_file(self) freq_from_seqs(self, seqs) freq_from_seqs_old(self, seqs) logbackground(self, seq) permute(self, letters, depth, seqs=[''], curdepth=0) study_seqs(self, seqs) zeroth(self)

Functions
		loadMarkovBackground(species='YEAST', seqs=[]) log2_sum(...) permute(depth, letters=['A', 'C', 'G', 'T'], seqs=[''], curdepth=0)

Data
		theMarkovBackground = None