invalid class “GRanges” object (#8b93668)

[ptrebert]

Hi David,

in the call for correctInvertedRegionPhasing, I get the following error:

    157   Working on chromosome cluster11
    158     calculating deltaWs ... 1.07s
    159     finding breakpoints ... 11.38s
    160     genotyping 1 ... 1.64s
    161     genotyping 2 ... 1.5s
    162     confidence intervals ... 2.63s
    163 Correcting phasing for inversions on chromosome: cluster11
    164 Error in validObject(.Object) : 
    165   invalid class “GRanges” object: 1: 'x@seqnames' is not parallel to 'x'
    166 invalid class “GRanges” object: 2: 'x@strand' is not parallel to 'x'
    167 Calls: correctInvertedRegionPhasing ... new_GRanges -> new -> initialize -> initialize -> validObject
    168 In addition: Warning messages:
    169 1: In parent.frame() :

I am copying the input data to the Globus share for debugging.
Best,
Peter

[ptrebert]

@daewoooo
re your email: for one, which version of StrandPhaseR did you run? another point: you are using a different parameter setting, see at least recall.phased (I guess lookup.bp has this default value as given below):

correctInvertedRegionPhasing(
    input.bams = bam.folder,
    outputfolder = output.folder,
    inv.bed = NULL,
    recall.phased = TRUE,
    het.genotype = 'strict',
    snv.positions = variant.calls,
    breakpointR.data = breakpointr.data,
    strandphaseR.data = strandphaser.data,
    pairedEndReads = TRUE,
    min.mapq = 10,
    background = 0.1,
    vcfs.files = strandphaser.vcf,
    lookup.bp = 1000000,
    ref.fasta = clustered.assm,
    assume.biallelic = TRUE
)

Apparently, I am not enough of an R programmer to understand how I have updated the documentation of the StrandPhaser function would affect the runtime behavior of a function?

[ptrebert]

@daewoooo what is the status here, could you reproduce this problem with StrandPhaseR #8b93668 ?

[daewoooo]

Hi Peter, I'm sorry I was attending a conference this week. I will resolve this issue tomorrow. David

[ptrebert]

No problem; it's the last sample that's causing trouble, the light at the end of the tunnel is probably not a train...

[ptrebert]

@daewoooo I am still waiting for feedback on this one