The Perl Toolchain Summit needs more sponsors. If your company depends on Perl, please support this very important event.
Revision history for diffReps.


1.55.3 Thu Sep 12 2013
    - diffReps will now remove intermediate files even if it dies from 
      exceptions.
    - The POD of diffReps modules are re-written so that they are more 
      informative.
    - Use Perl built-in functions to delete intermediate files instead of 
      external "rm" in the main program, making it more friendly to Windows 
      users.
    - Corrected some version numbers.


1.55.2 Sat Aug 31 2013
    - Corrected some mistakes in Makefile.PL.
    - Made the program aware of the number of required replicates when nb and 
      tt are selected.

1.55  Wed Jun 26 2013
    - Allow chromosome names to have more flexible format.
    - added example/norm.txt.

1.54  Fri Nov 2 2012
    - A quick-and-dirty implementation of parallel processing.
    - findHotspots now accepts option to specify columns to extract.
    - New default values for option "--nsd": broad(2) and sharp(20) peaks.
    - Column "logFC" is renamed to "log2FC" to avoid any ambiguity.

1.43  Mon Jul 23 2012
    - The null model used in findHotspot has been changed to local Poisson.
    - Added support for background samples, such as DNA input or IgG, to 
      calculate enrichment for each differential site.
    - Filter windows based on background enrichment (using this can make 
      detection >2 times slower).
    - Misc. improvement to make detecting differential sites more efficient 
      (~70% faster).

1.33  Tue Jun 19 2012
    - Output average read count for each site for further filtering purpose.
    - Output diffReps version and full command used.

1.32  Fri Mar 23 2012
    - Added script and database files for automatic annotation of differential
      sites.
    - Added script to look for chromatin modification hotspots from diffReps
      results.
      
1.12  Thu Mar 22 2012
    - Added robust estimation for background mean and deviation. Right-trimmed
      mean is used to replace arithmetic average. Median absolute deviation is
      used to replace standard deviation.
    - Robust estimation is done on a sub-sample of 100,000 bins instead of the 
      whole genome to save compuational cost and memory usage.

1.02  Thu Mar 8 2012
    - fixed a bug that will generate a warning message if you provide your own
      chromosome length file.
    - added a routine to choose window and step size more sensiblely and prevent
      weird values.

1.01  Fri Mar 2 2012
    - fixed a tiny bug that will generate a warning message if round-off errors
      cause the G-statistic to be a negative.
    - added scripts to convert from ELAND and SAM alignment to BED files.
    - added a script to remove redundant short reads.

1.00  Wed Feb 29 2012
    - a workable version created and tested.

0.01  Sat Feb 25 15:49:37 2012
    - original version; created by h2xs 1.23 with options
    - AX -b 5.6.0 -n diffReps