Hi.
On Tue, Oct 21, 2008 at 8:48 AM, yupu <[EMAIL PROTECTED]> wrote:
>
> Henrik:
>
> Hi, thanks for the response!
>
> Is there any way for me to find out all available APIs ( or functions)
> that come with the package?
>
> I tried ?getUnitsOnChromosome but there is not much information
> there :-S
I don't have the time to write detailed Rd help pages for end users
(and keep them up to date, which is the time-consuming part). I do
not know how many functions and methods there are, but we're talking
about more than 1000 pages of code. However, you can trust that the
usage/synopsis and the arguments are correct for Rd files. The Rd
help for a lot of the methods/functions are directed to a common page
(explaining it is still to be documented). The ones that exists are
mostly intended for developers, but developers tend to look at/follow
the source code instead.
Instead, my strategy is to document the package online. Some features
are also revealed in replies on the mailing list itself - so it's
worth skimming through message from others too. When a feature
discussed in the forum becomes frequently used, I try to write up a
page or add it to a vignette.
What is there you can trust. I am very conservative about what parts
of the API I make "public" this way, because for each new
method/feature I expose I will also have to keep backward compatible
which takes time.
The design of classes in aroma.affymetrix can be though of as classic
object-oriented programming as seem in C++, Java etc. This means that
you have an object of a certain class and there exist methods coupled
to that class. If you know the class, you can quickly get a list of
methods available, e.g.
> cdf <- AffymetrixCdfFile$byChipType("GenomeWideSNP_6")
> class(cdf)[1]
[1] "AffymetrixCdfFile"
# The Rd files for classes show an automatically generated list of all
(non-private) methods.
> help(AffymetrixCdfFile)
# You can get the same list by
> AffymetrixCdfFile
AffymetrixCdfFile extends UnitNamesFile, AromaChipTypeAnnotationFile, Affymetrix
File, AromaPlatformInterface, AromaMicroarrayDataFile, GenericDataFile, Object {
public convert(chipType=getChipType(this), suffix=NULL, sep="-", path="cdf", .
.., verbose=FALSE)
public createMonoCell(...)
public getCellIndices(units=NULL, ..., useNames=TRUE, unlist=FALSE, force=FALS
E, cache=TRUE, verbose=FALSE)
public getChipType(fullname=TRUE, ...)
public getDimension(...)
...[snip]
That may or not may be useful for advanced users.
As you work with the packages and maybe look at/follow the code
occasionally, you well learn the overall structure, design and sprit
of the package, which I hope will help out too. Posting questions
like yours is also a good for finding out about non-documented
features. If I see there are common use cases, I will not only try to
add methods to simplify them, but also document them online.
Conclusion: Use the aroma.affymetrix webpage/forum for documentation/help.
/Henrik
>
> Thanks!
> Yupu
>
>
>
> On Oct 20, 8:12 pm, "Henrik Bengtsson" <[EMAIL PROTECTED]> wrote:
>> Forgot one thing below:
>>
>>
>>
>> On Mon, Oct 20, 2008 at 5:02 PM, Henrik Bengtsson <[EMAIL PROTECTED]> wrote:
>> > Hi,
>>
>> > On Mon, Oct 20, 2008 at 3:51 PM, yupu <[EMAIL PROTECTED]> wrote:
>>
>> >> Hi,
>>
>> >> I am using aroma.affymetrix to do copy number estimation on
>> >> affymetrix's SNP6.0 data.
>>
>> >> I followed the instructions from
>> >>http://groups.google.com/group/aroma-affymetrix/web/total-copy-number...
>>
>> >> Things went smoothly, untill I realized there is no data point for at
>> >> position 51598 on Chr1 that is corresponding to probe id CN_473963.
>>
>> >>>units <- getUnitsOnChromosome(gi,chromosome=1)
>> >>>pos<- getPositions(gi,units=units)
>> >>>bool <- pos == 51598
>> >>> pos[bool]
>> >> integer(0)
>>
>> >> Did I do anything wrong? Or aroma is now dealing with these probes yet
>> >> for the moment?
>>
>> > Don't rely on exact genomic position in order to identify/refer to a
>> > CN variant/probe, because this information tend to get updated
>> > whenever the human genome annotation data is updated. Doing that is
>> > also error prone to differences in how positions are defined, e.g. if
>> > the first position of a chromsome is 0 or 1 (which might be issue in
>> > this case).
>>
>> > Instead, use the unit name to identify the position and not the other
>> > way around, e.g.
>>
>> >> cdf <- AffymetrixCdfFile$byChipType("GenomeWideSNP_6", tags="Full");
>> >> gi <- getGenomeInformation(cdf);
>> >> unit <- indexOf(cdf, name="CN_473963");
>> >> unit
>> > [1] 935591
>> >> pos <- getPositions(gi, units=unit);
>> >> pos
>> > [1] 51599
>>
>> > If you really want to identify a unit name given a position, allow for
>> > some shift, say +/-5 nucleotides, e.g.
>>
>> >> units <- getUnitsOnChromosome(gi, chromosome=1);
>> >> pos <- getPositions(gi, units=units);
>> >> idxs <- which(51598-5 <= pos & pos <= 51598+5);
>> >> units[idxs];
>> > [1] 935591
>> >> pos[idxs];
>> > [1] 51599
>>
>> A better way to identify units in a specific region on a given chromosome is:
>>
>>
>>
>> > units <- getUnitsOnChromosome(gi, chromosome=1, region=c(51598-5,
>> > 51598+5));
>> > units
>> [1] 935591
>> > pos <- getPositions(gi, units=units);
>> > pos
>> > [1] 51599
>>
>> I forgot about the 'region' arguments.
>>
>> /Henrik
>>
>>
>>
>> > Cheers
>>
>> > /Henrik
>>
>> >> Thanks,
>> >> Yupu
> >
>
--~--~---------~--~----~------------~-------~--~----~
When reporting problems on aroma.affymetrix, make sure 1) to run the latest
version of the package, 2) to report the output of sessionInfo() and
traceback(), and 3) to post a complete code example.
You received this message because you are subscribed to the Google Groups
"aroma.affymetrix" group.
To post to this group, send email to aroma-affymetrix@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at
http://groups.google.com/group/aroma-affymetrix?hl=en
-~----------~----~----~----~------~----~------~--~---
