noob questions regarding alignment of MiSeq data

[Azazel]

I will very soon have results from MiSeq and need to analyze them, in particular do the alignment. The samples have been prepared with a directional RNA-seq protocol, single end. Read length is up to ~300 bp. This is on homo sapiens.

I have a couple of very basic questions. Background: Usually I start my analysis from aligned sequences coming from HiSeq, I have no experience whatsoever with MiSeq. I am the "computer guy", and am not interacting with the instrument itself.

1. How does the output format of MiSeq look like? Is it FASTA or FASTQ or something else?
2. (main question) What is the best software for alignment? Can I use BWA? Do I need to perform putative PCR duplicate removal first, or removal of ribosomal sequences? Or can I just BWA the output and that's it? Anything else I need to be aware of, especially with respect to MiSeq?
3. The Illumina website alludes to something called "MiSeq Reporter" and states that it can do sequencing as well - is it worth pursuing?

Thanks!

[swbarnes2]

With RNA data, it's usually not a good idea to remove PCR duplicates, because very highly expressed genes will get whacked down in coverage. With single end data, especially, you will whack a lot of reads that are not actually duplicates.

You should probably just align to a reference genome that contains rRNA, that's the safest way to identify rRNA.

[westerman]

The miSeq reads are no different than the hiSeq data aside from being (a) longer @ 250+ bases and (b) less numerous. You can even mix the two without many hassles. If you are familiar with hiSeq work -- and this is unclear from your post -- then you should have no problems with the miSeq. Using directional reads, if you have not done so before, could be an interesting twist.

MiSeq reporter runs on the instrument computer. It might work with the human genome. I do not run it since 99.8% of our work is not human.

Best alignment software for RNA-seq on a mammalian organism? Probably not BWA (nor Bowtie2) because of the desire to skip the introns and map to the exons. Normally I'd recommend TopHat but there seems to be some problems with it lately (see other posts in SeqAnswers).