# 03/12/2014
# Changes made in the new (0.6) version of the TFTargetCaller package

1) "ensembl" changed to "geneID"
In the "gene" matrix the column containing gene IDs was named "ensembl". It was confusing because the user might wish to use other than Ensembl gene ID. In the current version "ensembl" has been changed to "geneID". All functions and documentation have been changed accordingly.

2) Gene strand might be provided either as (1, -1) or as (+, -)
Before the "strand" column in the "gene" matrix had to be numeric and contain "1" for forward strain and "-1" for reverse strand. It was confusing because some users might use "+" and "-". Now the possibility of using "+" and "-" is enabled (they are automatically translated to 1 and -1, respectively). In addition, if the symbols in the "stand" column are neither (1, -1) nor (+, -) an error message is displayed.

3) An internal error allegedly related to the use of the "structure" function was corrected.

4) An internal error making the function crash when for any chromosome there was just one gene in the "gene" matrix was corrected.

5) An argument "chromosomes" from the TFTargetCaller function has been removed. Before this argument needed to be used to specify the chromosomes that are to be taken into consideration. This might have been confusing, because the default was "c(1:19,"X")" for mouse and in order to analyze human data one would have had to change it to e.g. "c(1:22,"X")". This was explained in the documentation, however it was not very intuitive to use this argument. Now this is handled internally: the main function returns a score for every gene listed in the "genePosition" matrix. If some chromosomes are missing in the "peakPosition" table, the corresponding genes will just get a score of 0 (or 1, depending on the nature of the score).

6) Cosmetic text changes in the documentation files; contents not affected.