NAME:
RocSampler (Regularizing overlapping protein complexes) predicts protein complexes from weighted protein-protein interactions.

SYNOPSIS:
RocSampler inputFile outputFile OPTIONS

inputFile
	Specify a file of protein-protein interactions (PPIs) with
	thier weights.  Each line of the file should contain two
	protein IDs and the corresponding to the interaction weights,
	which are tab-deliminated.  PPIs without weights are also
	accepted. In this case, the weight of a PPI is set to be 1.

outputFile
	Specify an output file name which RocSampler makes. Each line of
	the output file corresponds a predicted cluster.

OPTIONS
	--ppi.norm <val>
		Determine whether PPI weights are normalized or not.
		<val> should be either 1 (default) or 0.

	--repeat-count <val>
		Set the repeat count of updating a set of clusters to
		be <val>.  The default value is 10,000,000.

	--cluster.num.max <val> 
		Set the upper bound of the size of a set of predicted
		clusters to be <val>. The default value is 1,000.

	--cluster.num.init <val> 
		Set the number of clusters in an initial set of
		predicted clusters to be <val>.  The default value is
		1.

	--temp.init <val>
		Set the value of an initial temperature to be <val>
		The default value is 1.

	--temp.init.repeat-count <val>
		RocSampler initially executes Metropolis-Hastings
		updates during <val> iterations with a specified
		initial temperature. The default value is 0.2.

	--temp.multiplier <val>
		After initial Metropolis-Hastings updates, the value
		of temperature is reduced from the previous one by
		multiplying it by <val>.  The default value is
		0.999999.

	--protein.multiplicity <val>
		Set the upper bound of the multiplicity of a protein
		to be <val>.  The default value is 2.

	--term.clu-den.alpha <val> 
		Set the exponent of the denominator of term,
		h_clu-den, to be <val>.  The default value is 0.5.

	--term.clu-size <val> 
		Determine whether term, h_clu-size, is set to be on or
		off.  <val> is either 1 (default) or 0.

	--term.clu-size.const <val>
		This parameter corresponds to the number of 1,000,000
		of the default setting of the coefficient of term,
		h_clu-size,s, in Table 1 of our paper. The default
		value is trivially 1,000,000.

	--term.clu-size.gamma <val>
		Set the value of gamma of term, h_clu-size,s, to be
		<val>.  The default value is 2.

	--term.pro-num <val> 
		Determine whether term, h_pro-num, is set to be on or
		off.  <val> is either 1 (default) or 0.

	--term.pro-num.target <val>
		Set the target number of proteins within clusters,
		which is used in term, h_pro-num, to be <val>.  The
		default value is 2000.

	--coef.pro-num <val> 
		Set coefficient, c_pro-num, to be <val>.  The default
		value is 0.005.

	--term.clu-dis <val> 
		Determine whether term, h_clu-dis, is set to be on or
		off.  <val> is either 1 (default) or 0.

	--term.clu-dis.beta <val> 
		Set the value of beta of term, h_clu-dis, to be <val>.
		The default value is 0.4.

	--coef.clu-dis <val>
		Set coefficient, c_clu-dis, to be <val>.  The default
		value is 10.

	--term.clu-num <val> 
		Determine whether term, h_clu-num, is set to be on or
		off.  <val> is either 1 (default) or 0.

	--term.clu-num.target <val> 
		Set the target number of clusters, which is used in
		term, h_clu-num, to be <val>.  The default value is
		400.

	--coef.clu-num <val> 
		Set coefficient, c_clu-num, to be <val>.  The default
		value is 0.005.

	--term.over-num <val> 
		Determine whether term, h_over-num, is set to be on or
		off.  <val> is either 1 (default) or 0.

	--coef.over-num <val> 
		Set coefficient, c_over-num, to be <val>.  The default
		value is 0.005.

	--term.clu-int <val> 
		Determine whether term, h_clu-int, is set to be on or
		off. <val> is either 1 (default) or 0.

	--coef.clu-int <val> 
		Set coefficient, c_clu-int, to be <val>.  The default
		value is 0.01.

	--proposal.alphaac <val> 
		Set probability, alpha_ac, to be <val>.  The default
		value is 0.25.

	--proposal.alphaap <val> 
		Set probability, alpha_ap, to be <val>.  The default
		value is 0.25.

	--proposal.alpharc <val> 
		Set probability, alpha_rc, to be <val>.  The default
		value is 0.25.

	--score.plot <val>
		Determine wheter a graph of all score values during
		iterations is drawn or not.  <val> is either 1
		(default) or 0.

	--score.verbose <val>
		 Determine whether a verbose mode, in which the
		 current term scores are displayed, or not. <val> is
		 either 0 (default) or 1.


EXAMPLE:
	See file, How-to-use.txt.