In Version 0.1, a faster method is used to calculate the GC content in DtaPrep.c.
		fixed a bug in printing the expected value

In Version 0.1.1, modified the program to remove the reads near the gap of the genome assembly (BRS program)

In Version 0.1.2, this program can be used to ONLY bin the data

In Version 0.1.3, added perl scripts to remove the temp files generated.
		  modified BRS program to report the total number of reads on mappable positions and total number of mappable positions.
		  estimate the bin size used for the refine step
		  modified the count_nuc function in DtaPrep.c
		  also modified the function calculate_bias in DtaPrep.c allowing read positions less than fragment size and read length.
		  changed the default bin type to be bins with equal number uniquely mappable positions


In Version 0.1.4, suppressed reporting the ratio between the observed and expected

In Version 0.1.5, provide functions to only bin the data and plot the read count vs gc content

In Version 0.1.6, first iteratively remove the regions that are likely to be CNV regions (use linear regression)
		  In the earlier version, the last bin was processed differently for the bin only procedure and the prediction procedure;
		  this is fixed in this version 
		  In this version, some temp files will not be deleted.

In Version 0.1.6.1, fixed some bug that may cause the program freezing (a bug in read.c)
		  the output contains more information
		  Fixed a bug in the code normalized.pl (a bug in checking if the config file has correct information)

In Version 0.1.7, allow to extend the read to upstream and downstream for model training and fitting.

In version 0.1.7.1, Fixed a bug in DtaPrep.c
		    Only consider nucleotide composition near the start and end of a read

In version 0.1.7.2, removed the function about the gap of the genome assembly (only the perl program)
		    Fixed a but in compRatio.R to avoid accasional negative prediction


In version 0.2, use nonparametric regression instead of polynomial regression (glm step)


In version 0.2.1, use nonparametric regression in the refine step
		  changed pos = pos_in - 1 to pos = pos_in in DtaPrep.c and DtaPrepGAM.c
                plan to seperate the nonparametric normalization procedure and the parametric procedure as seperate perl pipeline (Didn't work on this yet)

In version 0.2.2, add a function such that the users can choose to not filter the potential CNV regions
		  removed some unused options, such as gc_mp, gc_log, gc_degree, gconly, gapfile
		
In version 0.2.3, add one step in purityEM to remove outliers



