[Bioc-devel] Bug in frmaTools

I've also encoundered another problem, this time I believe related to a 
change the the frma package which requires a corresponding change in the 
frmaTools package. Apparently frma has added a check for equality of 
probe names in various places, and frmaTools does not store these names, 
which means that the current version of frma will reject any vectors 
generated by the current version of frmaTools with the following error:

 > eset <- frma(affy[,1:3] , input.vecs=myFrmaToolsVecs)
Error in frmaAffyBatch(object, background, normalize, summarize, 
input.vecs,  :
   Mismatch between pmindex(object) and names of input.vecs and unable 
to create unique mapping.
 > traceback()
3: stop("Mismatch between pmindex(object) and names of input.vecs and 
unable to create unique mapping.")
2: frmaAffyBatch(object, background, normalize, summarize, input.vecs,
        output.param, verbose)
1: frma(affy[, 1:3], input.vecs = myFrmaToolsVecs)

 > sessionInfo()
R version 3.2.0 (2015-04-16)
Platform: x86_64-unknown-linux-gnu (64-bit)
Running under: Ubuntu 15.04

locale:
  [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C
  [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8
  [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8
  [7] LC_PAPER=en_US.UTF-8       LC_NAME=C
  [9] LC_ADDRESS=C               LC_TELEPHONE=C
[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C

attached base packages:
[1] grDevices datasets  parallel  graphics  stats4    stats utils
[8] methods   base

other attached packages:
  [1] frmaTools_1.20.0
  [2] hgu133afrmavecs_1.5.0
  [3] hthgu133pluspmcdf_2.16.0
  [4] DSalomon.PAX.hthgu133pluspmfrmavecs_0.1
  [5] frma_1.20.0
  [6] affy_1.46.1
  [7] Biobase_2.28.0
  [8] openxlsx_2.5.10
  [9] foreach_1.4.2
[10] plyr_1.8.3
[11] stringr_1.0.0
[12] IRanges_2.2.4
[13] ggplot2_1.0.1
[14] S4Vectors_0.6.0
[15] BiocGenerics_0.14.0
[16] BiocInstaller_1.18.3

loaded via a namespace (and not attached):
  [1] Rcpp_0.11.6           compiler_3.2.0 GenomeInfoDb_1.4.1
  [4] XVector_0.8.0         iterators_1.0.7 tools_3.2.0
  [7] zlibbioc_1.14.0       digest_0.6.8 bit_1.1-12
[10] RSQLite_1.0.0         preprocessCore_1.30.0 gtable_0.1.2
[13] ff_2.2-13             DBI_0.3.1 proto_0.3-10
[16] affxparser_1.40.0     Biostrings_2.36.1 grid_3.2.0
[19] AnnotationDbi_1.30.1  oligo_1.32.0 reshape2_1.4.1
[22] magrittr_1.5          splines_3.2.0 scales_0.2.5
[25] codetools_0.2-11      oligoClasses_1.30.0 MASS_7.3-41
[28] GenomicRanges_1.20.5  colorspace_1.2-6 stringi_0.5-2
[31] munsell_0.4.2         affyio_1.36.0

Ryan,

Thanks for pointing these out. I'll look into them soon, but I imagine 
your assessment is correct.

Best,
Matt



On Mon, Jul 6, 2015 at 5:42 PM, Ryan C. Thompson <rct at thompsonclan.org 
<mailto:rct at thompsonclan.org>> wrote:

    I also discovered another apparent bug later in the same function.
    The second to last line of makeVectorsAffyBatch is

        vers <- ifelse(!is.null(cdfname),
    as.character(packageVersion(cdfname)), "")

    If cdfname is NULL, this line will throw an error because the
    second argument to ifelse will have length zero and "ifelse" does
    NOT do lazy evaluation.


    On 07/06/2015 12:21 PM, Ryan C. Thompson wrote:

        Hello,

        I just encountered a bug in frmaTools that makes it impossible
        to use on certain array platforms. The following lines in
        makeVectorsAffyBatch fail on an AffyBatch object on the
        hthgu133pluspm platform:

            pms <- pm(object)
            pns <- probeNames(object)
            pmi <- unlist(pmindex(object))
            if (!identical(as.character(pmi), rownames(pms)))
                stop("Mismatch between pmindex and rownames of pms")

        I isolated the problem to five probes:

        > i <- which(as.character(pmi) != rownames(pms))
        > pmi[i]

         1564498_PM_at9 205398_PM_s_at8 217695_PM_x_at7
        223446_PM_s_at7 237802_PM_at3
                  3e+05           5e+05           2e+05 1e+05 4e+05

        > rownames(pms)[i]

        [1] "300000" "500000" "200000" "100000" "400000"

        > as.character(pmi)[i]

        [1] "3e+05" "5e+05" "2e+05" "1e+05" "4e+05"

        As you can see, the problem is that as.character will happily
        use scientific notation when it feels like it, which then
        fails a test for string equality. I believe the solution is to
        replace that test with:

        all(sprintf("%i", pmi) == rownames(pms))

        -Ryan

    _______________________________________________
    Bioc-devel at r-project.org <mailto:Bioc-devel at r-project.org> mailing
    list
    https://stat.ethz.ch/mailman/listinfo/bioc-devel




-- 
Matthew N McCall, PhD
6 Bridgewood Dr.
Fairport, NY 14450
Cell: 202-222-5880