Biomedical Engineering Reference
In-Depth Information
workloads we must always pick our battles and sadly the extent of our
contribution to Open Source projects is often little more than to report
bugs to the project owner. A saving grace is that sometimes we are able
to become part of the wider development community and have some
infl uence on the way that software we use gets made or just get inside
information on how to use it and how to code with it. Such contacts are
worth their weight in gold.
Given such issues, why would we not use commercial software? If you
pay, surely this results in better support? The answer is sometimes true,
with commercial software providing the only practical solution. Usually
this will be the case when buying a new piece of hardware, we are stuck
with the vendor's software for running all of our sequencing machines,
job management systems and storage devices. There is no open source
alternative software that could be relied on for these absolutely critical
parts of the pipeline. For the downstream analyses, though, the simple
fact is that commercial software is rarely up to the task. In aiming for the
largest markets, commercial tools often try to incorporate too many
functions. This can often result in the tool being 'jack of all trades, master
of none', with results being unsatisfactory or too much of a black box for
scientists to be comfortable with. Tying users in to proprietary standards
is another annoyance, legacy data in undocumented or proprietary
formats that no other software reads makes it impossible to re-analyse in
other packages. Science moves fast and new methods appear all the time,
commercial software seems rarely to innovate.
Ultimately, this need for speed is why free and open source software will
always be a vital component of the software pipelines of most scientifi c
research institutes. The need of the community to innovate could never
really be matched by commercial efforts that need to provide solidity and
make a return on their investment. As painful as the time and meddling
aspect of using open source is, it does help us to keep as current as we need
to be. The best middle ground will come from open source projects like
Galaxy and Taverna [33], and standard data format descriptions like
SAM [34] or Omero-TIFF [15], that help us to tie disparate bits together
and make both interoperability and usability of the rapidly appearing and
evolving tools much greater where it really counts, with the scientist.
11.11 References
[1] Moore's Law -
http://en.wikipedia.org/wiki/Moore's_law
[2] Illumina Hi Seq -
http://www.illumina.com/systems/hiseq_2000.ilmn
Search WWH ::
Custom Search