"laminaria.EST" wrote:
>> Brian Fristensky wrote :
>> Whoa -- I guess I missed something. What is a ?Method object, and
> can these be defined within ACeDB or Blixem? Where are ?Method
> objects documented?
> ---------------------------------
>> Hello
>> What I have understood is that ?Method object correspond to the display definition. Following is a copy of the ?Method as defined in the ACEDB 4_7g model. I do not know where it is documented. If somebody knows, I am interested.
>> ?Method Remark ?Text
> // the Display information controls how the column looks.
> Display Colour #Colour
> Frame_sensitive
> Strand_sensitive Show_up_strand #Colour
> Score Score_by_offset // has priority over width, for Jean
> Score_by_width
> Score_by_histogram UNIQUE Float // baseline value
> Score_bounds UNIQUE Float UNIQUE Float
> // limits for squashing
> Percent // '%' in text, + default bounds 25, 100
> Overlap_mode UNIQUE Overlap // draw on top - default
> Bumpable // bump to avoid overlap
> Cluster // one column per homol target
> Width UNIQUE Float
> Symbol UNIQUE Text // first char only is used
> Right_priority UNIQUE Float // the greater the more right
> Max_mag UNIQUE Float // don't show if more bases per line
> Min_mag UNIQUE Float // don't show if fewer bases per line
> Show_text // only put text in right-hand column if set
> Blastn // can calculate percent from score if blastn
> Blixem Blixem_X
> Blixem_N
> Blixem_P
> Belvu
> EMBL_dump EMBL_feature UNIQUE Text // require this
> EMBL_threshold UNIQUE Float // apply to score unless overridden
> EMBL_qualifier Text UNIQUE Text
> // if 1 Text, it is the entire qualifier including '/'
> // if 2 Texts, 1st is an sprintf format and 2nd is
> // an argument. If this is "score", "note", "y1", "y2" or "target"
> // then use the corresponding field of the Feature or Homol line.
> // multiple formats will be concatenated until one starts with '/'.
> Gene_assemble Gene_assemble_method ?Method UNIQUE Float
> // Float factor to multiply score by
> Intron_min UNIQUE Int
> Exon_min UNIQUE Int
> Intron_cost UNIQUE Float UNIQUE Float UNIQUE Int
> // base cost, cost per log bp beyond min
> // Int is min for this if not Intron_min
> Inter_gene_cost UNIQUE Float
> GF_parameters GF_range UNIQUE Int
> // +- range for genefinder feature calculation
> GF_ATG_cutoff UNIQUE Float
> GF_5_cutoff UNIQUE Float
> GF_3_cutoff UNIQUE Float
> GFF GFF_source UNIQUE Text
> GFF_feature UNIQUE Text
> EMBL_dump_info #EMBL_dump_info // when used for controlling dump
--
So, this object, whose name is Method, is used for displaying output
from various
programs such as BLAST, BLIXEM, GENEFINDER etc. This raises several new
points:
1. I get the feeling that I still don't have the entire story on
?Method.
For example, in the ACeDB models.wrm file listed at
http://probe.nalusda.gov:8000/acedocs/models/models.html
has a very different object called ?Models, so I guess that
the above object definition is specific to a particular
ACeDB-based database.
Nonetheless, this definition seems to imply that output from
various programs is being displayed by ACeDB. Presumably,
that is done by Pick_me_to_call from some other object?
Or is there a mechanism for xace to display the output?
2. At the risk of being impertinent, in my humble opinion,
it is not a good idea to call an object by the name "Method",
because the word "method" has a very precise, and important
meaning in Object-Oriented Programming.
3. On that subject, here is a probably naive question.
As an idea for the future evolution of ACeDB, is it
plausible for objects to have methods, in the OOP
sense? This could be implemented in a very limited
way, much along the lines of how 'Pick_me_to_call'
is implemented now. That is, as a way for calling
external programs, or displaying files or output from
programs. Is it possible to get even more sophisticated,
such as interfacing with CORBA objects and their methods?
I don't think we want to see xace itself become
bloatware. Rather, I like the idea of being able to
use external objects and methods to extend functionality,
in a manner analogous to how GDE can run most types
of sequence analysis programs.
===============================================================================
Brian Fristensky | Original entry for Microsoft Windows,
Department of Plant Science | in the Hitchiker's Guide to
University of Manitoba | the Galaxy: "Harmless".
Winnipeg, MB R3T 2N2 CANADA |
frist at cc.umanitoba.ca | After the release of Windows NT, this
Office phone: 204-474-6085 | was changed to "Mostly harmless".
FAX: 204-261-5732 |
http://home.cc.umanitoba.ca/~frist/
===============================================================================
