[mira_talk] Re: RE reading fasta files
- From: Bastien Chevreux <bach@xxxxxxxxxxxx>
- To: mira_talk@xxxxxxxxxxxxx
- Date: Fri, 3 Apr 2009 00:03:19 +0200
On Thursday 02 April 2009 Jan Paces wrote:
> Jorge.DUARTE@xxxxxxxxxxxx wrote:
> > I had a similar problem myself, and changing the default value for
> > option -AS:bdq solved it.
> > SANGER_SETTINGS -AS:bdq=40
> Thanks, works now.
Thanks Jorge for answering as I had been on a business trip without much
internet access.
The above problem has popped a number of times and I'm struggling with myself
on how to configure MIRA best so that people get what they expect ...
Problem is: normally, an assembler should have qualities given to work well.
On the other hand, some data sets in hybrid assemblies simply don't have any
data which can be given.
So, if a user now gives MIRA data with qualities (as should normally be the
case), but the clipping is weak or non-existent, then all hell breaks loose if
there is bad quality remaining in the data and one gets sub-optimal assemblies
(in the best case) or simply a whole lot of small contigs (leading to "MIRA
does not perform!" mails *sigh*).
In earlier times MIRA insisted to have qualities or being explicitly told via
parameters that there would be no qualities for a given data set. Which led to
a number of questions like "I don't have qualities, how can I still use MIRA?"
when users didn't find the correct parameter (and the number of these has grown
since).
So, at the moment, I settled for "if there are no qualities, the user has to
know what he's doing and sets flags accordingly.
I'd be happy to hear suggestions for a better handling though. I personally am
a big fan of "if it does not meet the expectations of the program, fail".
Regards,
Bastien
--
You have received this mail because you are subscribed to the mira_talk mailing
list. For information on how to subscribe or unsubscribe, please visit
http://www.chevreux.org/mira_mailinglists.html
Other related posts:
