[Bioc-devel] sequencing workflow

From: "Michael Lawrence" <lawrence.michael at gene.com>
To: bioc-devel at r-project.org
Sent: Tuesday, December 17, 2013 4:26:20 PM
Subject: [Bioc-devel] sequencing workflow

Hey guys,

I was just checking out the workflows on the Bioc website, putting
myself
in the mindset of a naive user. At first glance, the sequencing
workflow is
pretty intimidating. There's a lot of code and text in a _tiny_ font,
the
examples do not run out of the box, and they expect the user to
create a
particular directory layout just to get started.

What about starting with a cookbook-style tour of the neat things one
can
do with ranges and sequences, and then integrate everything into a
complete
workflow at the end? With sample data made immediately available. For
example, while running a ShortRead QA report is the right thing to do
when
starting from a FASTQ, it might not be the best thing for selling the
user
on the richness of the environment, and the coolness of how the tools
interoperate. Just some thoughts. Not volunteering ;)

An important thing to keep in mind is that if the user tries to run
the
examples on a full-scale dataset, they are only going to succeed in
crashing their machine. So either we need to teach them how to write
scalable code, or we need to provide high-level functionality that
handles
that for them, until they are ready to dig deeper.

Michael

	[[alternative HTML version deleted]]