A forward-backward fragment assembling algorithm for the identification of genomic amplification and deletion breakpoints using high-density single nucleotide polymorphism (SNP) array

Table 1 Parameters tested for the seven R packages

Packages		Parameters tested
Tuned packages		Tuning parameter	Values tested	Other parameters
	FASeg	Sig	0.25, 0.1, 0.075, 0.05, 0.025, 0.01, 0.005, 0.001, 0.0001, 0.00001	Default
	aCGH	Vr	10, 7, 5, 2, 1, 0.5, 0.1, 0.05, 0.01, 0.001	Default
	DNAcopy	alpha	0.25, 0.1, 0.075, 0.05, 0.025, 0.01, 0.005, 0.001,0.0005, 0.0001	* nperm = 1000
	GLAD	qlambda	0.75, 0.9, 0.925, 0.95, 0.975, 0.99, 0.9925, 0.995, 0.9975, 0.999	** lambdabreak = 0.01 lambdacluster = 0.01 lambdaclusterGen = 0.01 param = c(d = 1)
Packages examined at default setting	Picard	Maxk = max(true segment size) + 5, maxSeg= #(true segments) + 1
	RJaCGH	* burnin = 50, * TOT = 500, jump.parameters = NULL, k.max = #(true states) + 1
	BioHMM	Default

* The change in the number of permutations is to reduce computing time. Experiments showed that reducing the number from 10000 to 1000 has minimal effect on the outcome.
** These parameters were tuned according to the GLAD manual to increase sensitivity. Using default values, the method detected limited number of edges from noisy data.
*** The purpose of reducing the number of iterations was to save computing time.

ISSN: 1471-2105