[BioC] Please Help with rMAT: NAs in Enrich Object

[Martin Morgan]

Hi Noah --

Noah Dowell wrote:
> Hello All,
> 
> I am trying to use the rMAT package to analyze Affymetrix tiling array chip-chip data but have run into some problems that I hope the board can help with.
> 
> Up to normalization step looks fine:
> 
>> summary(wtAsetNorm)
> 
>    Genome interrogated:  Sc03b_MR_v04  
>    Chromosome(s) interrogated: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, mitochondrion, plasmid 
>    Sample name(s):  WTa WTb input1 input2  
>    The total number of probes is:  2643812  
>    Preprocessing Information 
>      - Transformation: log 
>      - Normalization: MAT standardized 
> 
> #This is what I used to find enriched regions:
> 
>> Enrich  <- callEnrichedRegions(RD, dMax = 75, dMerge = 200, nProbesMin = 8, method = "score", threshold = 1, verbose = FALSE)
> 
> #Then after loading the rtracklayer package I start to get some errors that I pasted below:
> 
>> genome(Enrich) <- "sacCer2"
>> names(Enrich) <- "chrI"

Perhaps it is related to

> x=RangedData(IRanges(1, 1:3), space=c(1,1,2))
> names(x) = "chrI"
> x
RangedData with 3 rows and 0 value columns across 2 spaces
        space    ranges |
  <character> <IRanges> |
1        chrI    [1, 1] |
2        chrI    [1, 2] |
3        <NA>    [1, 3] |

which introduces NA into 'space'. Maybe (though not quite -- mitochrion
and plasmid need to be renamed or removed)

> x=RangedData(IRanges(1, 1:3), space=c(1,1,2))
> names(x) = paste("chr", as.roman(names(x)), sep="")
> x
RangedData with 3 rows and 0 value columns across 2 spaces
        space    ranges |
  <character> <IRanges> |
1        chrI    [1, 1] |
2        chrI    [1, 2] |
3       chrII    [1, 3] |

? Martin

>> session <- browserSession("UCSC")
>> track(session, "toto") <- Enrich
> Error in ucscNormSeqNames(names(from)) : 
>   All sequence names should begin with 'chr' or 'scaffold'
>> subEnrich <- Enrich[2, ]
>> view <- browserView(session, range(subEnrich) * -2)
> Error in asMethod(object) : 
>   cannot create an IRanges object from an integer vector with missing values
> In addition: Warning messages:
> 1: In min(x, na.rm = na.rm) :
>   no non-missing arguments to min; returning Inf
> 2: In max(x, na.rm = na.rm) :
>   no non-missing arguments to max; returning -Inf
> 3: In .class1(object) : NAs introduced by coercion
> Error in ucscForm(range) : 
>   error in evaluating the argument 'object' in selecting a method for function 'ucscForm'
> 
> #the top of the file looks fine but looking at the tail we see part of the problem:
>> tail (Enrich)
> RangedData with 6 rows and 1 value column across 18 spaces
>         space         ranges |     score
>   <character>      <IRanges> | <numeric>
> 1        <NA> [61862, 61934] | 2.0867882
> 2        <NA> [63373, 63377] | 1.2274179
> 3        <NA> [74032, 74443] | 1.3762046
> 4        <NA> [79253, 80141] | 1.6481782
> 5        <NA> [80710, 80710] | 0.5607601
> 6        <NA> [    0,  6261] | 5.9487499
> 
> 
> 
> Question:  Where did these NAs come from?  I tried to grep on "chrI" to create a subset lacking the NAs but I think this is a fundamental problem with something I am doing.  Any advice would be appreciated.  Thank you in advance.
> 
> 
> 
> Noah Dowell
> _______________________________________________
> 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


-- 
Martin Morgan
Computational Biology / Fred Hutchinson Cancer Research Center
1100 Fairview Ave. N.
PO Box 19024 Seattle, WA 98109

Location: Arnold Building M1 B861
Phone: (206) 667-2793