[R] Merging listed dataset into one

[David Winsemius]

On Jul 1, 2012, at 6:15 AM, Rui Barradas wrote:

> Hello,
>
> That error is because your datum is of a class that merge.default  
> cannot handle, like the error message says. By the look of it, class  
> "TillingFeatureSet" has a print method written to print it. This  
> because it prints objects of that class with some descriptive  
> information not part of the object itself. The underlying structure  
> seems to be a list, which was subclassed to a "TillingFeatureSet"  
> class object, with methods specially written for it.

Actually the error was from the coercion function  
`as.data.frame.default`. This rawData object appears to be a list of  
two S4 objects, (but that is a guess) and I suspect it was created by  
a BioConductor package. It's not surprising that a complex irregular  
object would not have an "as" method.  (None of the requested  
information about the package being used (very possibly "oligo") or  
other system details were provided.)

The question should probably be reformulated with more details and  
sent to the BioConductor mailing list.

-- 
David.

>
> Bottom line: if you merge the two components, the special behavior  
> that comes with its class might be lost. Do you really need to merge  
> them?
>
> If so, use dput(rawData), its output, to post your object. But think  
> about it first.
>
> Hope this helps,
>
> Rui Barradas
>
> Em 28-06-2012 16:47, Seungyeul Yoo escreveu:
>> Hello,
>>
>> I'm wondering how I can merge two featuresets into one.
>>
>> My dataset is two sets of microarray data and it looks like  
>> followings:
>>
>>> rawData
>> $v1
>> TilingFeatureSet (storageMode: lockedEnvironment)
>> assayData: 2197815 features, 59 samples
>>   element names: channel1, channel2
>> protocolData
>>   rowNames: LT290677RU_D1_2011-02-16 LT286300LU_D1_2010-07-24 ...
>>     LT003990RU_D1_2010-11-04 (59 total)
>>   varLabels: filenamesChannel1 filenamesChannel2 dates1 dates2
>>   varMetadata: labelDescription channel
>> phenoData
>>   rowNames: LT290677RU_D1_2011-02-16 LT286300LU_D1_2010-07-24 ...
>>     LT003990RU_D1_2010-11-04 (59 total)
>>   varLabels: sampleID tissue ... Annotation (5 total)
>>   varMetadata: labelDescription channel
>> featureData: none
>> experimentData: use 'experimentData(object)'
>> Annotation: pd.feinberg.hg18.me.hx1
>>
>> $v1.1
>> TilingFeatureSet (storageMode: lockedEnvironment)
>> assayData: 2197815 features, 17 samples
>>   element names: channel1, channel2
>> protocolData
>>   rowNames: LT282562RM_D1_2010-11-22 LT280646RU_D1_2010-11-22 ...
>>     LT093297LU_D1_2010-11-12 (17 total)
>>   varLabels: filenamesChannel1 filenamesChannel2 dates1 dates2
>>   varMetadata: labelDescription channel
>> phenoData
>>   rowNames: LT282562RM_D1_2010-11-22 LT280646RU_D1_2010-11-22 ...
>>     LT093297LU_D1_2010-11-12 (17 total)
>>   varLabels: sampleID tissue ... Annotation (5 total)
>>   varMetadata: labelDescription channel
>> featureData: none
>> experimentData: use 'experimentData(object)'
>> Annotation: pd.feinberg.hg18.me.hx1
>>
>> I want to merge rawData$v1 and rawData$v1.1 since they are  
>> consistent each other.
>>
>> I was trying to use "merge" function as merge(rawData$v1, rawData 
>> $v1.1) but it failed with error as
>>
>> Error in as.data.frame.default(x) :
>>   cannot coerce class 'structure("TilingFeatureSet", package =  
>> "oligoClasses")' into a data.frame
>> Calls: merge ... merge.default -> merge -> as.data.frame ->  
>> as.data.frame.default
>>
>> Can you please help with some advices?
>>
>> Thanks,
>>
>> Best regards,
>>
>> Seungyeul Yoo
>>
>> Postdoc Fellow,
>> Institute of Genomics and Multiscale Biology
>


David Winsemius, MD
West Hartford, CT