[BioC] Subsetting data from RGList (Again)

[Martin Morgan]

Elliott

Great! some progress. I'm not sure where your problem is coming from
now, but suspect that you've managed to assign something to 'R' that
is of the wrong sort. What is class(R)? names(R)? sapply(R, dim)? Can
you recreate R?  When you try to subset and see an error, what does
traceback() say?  And so that we are on the same page, what does
sessionInfo() say?

Martin

"elliott harrison" <e.harrison at epistem.co.uk> writes:

> Thanks Martin,
>
> I'd sorted the which removal and get a sensible matrix of matches
>
>> idx <- match(a$Gene.Probe, R$genes$ProbeName)
>> idx
>   [1]    38    65   200   278   303   312   350   351   388   571   710
> 714   778   810   997  1027  1049  1070  1177  1229  1245  1307  1328
> 1338  1411
>
>
> So I'm finding probes in my data set Hurrah!!!
>
> However when I try to apply it
>
>> R[idx,]
> Error in R[idx, ] : incorrect number of dimensions 
>
> Any idea what is wrong here?
>
> Thanks again
>
> Ellliott
>
> -----Original Message-----
> From: Martin Morgan [mailto:mtmorgan at fhcrc.org] 
> Sent: Tuesday, October 30, 2007 3:50 PM
> To: elliott harrison
> Cc: bioconductor at stat.math.ethz.ch
> Subject: Re: [BioC] Subsetting data from RGList (Again)
>
> Hi Elliott,
>
> "elliott harrison" <e.harrison at epistem.co.uk> writes:
>
>> Hi All,
>> Sorry to trouble everyone  but I didn't get my subsetting of an RGList
>
>> sorted last week.
>> I've tried loads of different combinations of things that were 
>> suggested but not quite there yet.
>>
>> MY Data is an RGList object here is a bit
> [SNIP]
>> $genes
>>   Row Col FeatureNum ProbeUID ControlType       ProbeName
>> GeneName  SystematicName   Status
>> 1   1   1          1        0           1 GE_BrightCorner
>> GE_BrightCorner GE_BrightCorner Control6
>>  
>> $source
>> [1] "generic"
>>  
>>  
>> My list of probes to select data with is in this object.
>>> a
>>            Gene.Probe
>> Probe1   A_24_P329635
>> Probe2   A_23_P104098
>> Probe3   A_23_P419038
>> Probe4   A_24_P124550
>> Probe5    A_23_P99044
>   
> You'd like to match your 'Gene.Probe' column, I guess, to a column in
> R$genes, maybe ProbeName? So create a numeric vector, with the locations
> of each Gene.Probe in R$Genes$ProbeName
>
>> idx <- which(match(a$Gene.Probe, R$genes$ProbeName))
>
> and use it to subset R
>
>> R[idx,]
>
> If you'd wanted to match Probe1, etc, you'd have used rownames(a)
> instead of a$Gene.Probe. Sometimes RGList has row names, i.e.,
>
>> rownames(R)
>
> returns a (long!) character vector. In these cases, you can create a
> character vector 'rnms' containing at least some of the row names of R.
> and
>
>> R[rnms,]
>
> Martin
>
>>  
>>  
>> I've tried
>> A=rownames(a)
>> RA = R[A,]
>>  
>> But don't get any genes
>>>$genes
>> [1] Row            Col            FeatureNum     ProbeUID
>> ControlType    ProbeName      GeneName       SystematicName Status
>>
>> <0 rows> (or 0-length row.names)
>>  
>> I've tried creating a vector using match a load of different ways but 
>> it is always empty
>>>vec = match(A, R)
>>>vec = match(A, R$genes$ProbeName)
>>>vec = match(A, R$genes)
>>>vec = match(probelist600, R)
>>>vec = match(probelist600, R$genes)
>>>vec = match(probelist600, R$genes$ProbeName)
>>  
>>  
>> This is part what ends up in object A
>>> A
>>   [1] "Probe1"   "Probe2"   "Probe3"   "Probe4"   "Probe5"   "Probe6"
>> "Probe7"   "Probe8"   "Probe9"   "Probe10"  "Probe11"  "Probe12"
>> "Probe13" 
>>
>> I expected to see my probe names in there.
>> I'm clearly not getting something. 
>> Can anyone see where I'm going wrong?
>> Many Thanks
>> Elliott
>>
>>
>> This message has been scanned for viruses by 
>> BlackSpider...{{dropped:3}}
>>
>> _______________________________________________
>> Bioconductor mailing list
>> Bioconductor at stat.math.ethz.ch
>> https://stat.ethz.ch/mailman/listinfo/bioconductor
>> Search the archives: 
>> http://news.gmane.org/gmane.science.biology.informatics.conductor
>
>
>  
>
>
> This message has been scanned for viruses by BlackSpider MailControl - www.blackspider.com