Need to figure out how to work with the references in the document. Might try the Pandoc way, but have also seen a proposal for an R package.

Need to return both the actual TMmm intervals, as well as the merged Exon intervals, and the original exon intervals returned

I want to create a markdown version of the submitted paper, and host this on the gh-pages branch of the repo.

This will require a few things:

Rewriting the output so that we save the alignments for the mm and pm probes, and their overlapping exons from merged exons and base exons
Basically I want to be able to do the full paper generation from the output text files, or the RData file with all of the summary of results.

Saving the actual intensity difference matrix so we can refer back to it

Refactoring all of the big code chunks in the lapply into sub-functions so that we don't have to worry about deleting objects after we are done with them.

Figure out how generate a pdf using pandoc and pdflatex. Currently we get an error from pdflatex when trying to generate the pdf. Not sure why.

Using the paper as a guideline, create a presentation for bioinformatics journal club next week. I really want to use markdown slides, so figure out how to do it.

So in practicing the presentation the night before I give the talk at Biomedcom, I realize that perhaps we did not do the analysis of non-unique PM alignment correctly. The current analysis looks at it by simply the numbers of PMs from each class that had more than one alignment to the genome, and the percentages are then calculated based on the total of probes that aligned multiply.

However, what I realized we should do is take it as a percentage of the total number of probes on the array in that class. For example:

AT should be:

of at probes that multiply align / # of at probes on the chip

with the current data, how do we count the set of mm probes that align to all 5 human builds that overlap with exons only?

The problem with the current data is that the probeAlign that is saved, is only based on those probes that align once in the genome, not all alignments. But maybe that will be good enough ...

Where appropriate, use the svg versions of the figures