Hello everyone!

I want to analyze my PAR-CLIP data utilizing wavClusteR. So when I try to create count table from my bam file (prepared by CLC WB) by command below

countTable <- getAllSub( Bam, minCov = 10 )

I have the error like that:

Error in { : 
  task 1 failed - "solving row 140: range cannot be determined from the supplied arguments (too many NAs)"

My bam:

> Bam
GRanges object with 872714 ranges and 2 metadata columns:
             seqnames           ranges strand |                    qseq          MD
                <Rle>        <IRanges>  <Rle> |          <DNAStringSet> <character>
       [1]       chr1 [ 57283,  57298]      + |        CAGTCATTCCGAACAA          16
       [2]       chr1 [ 57283,  57298]      - |        CAGTCATTCCGAACAA          16
       [3]       chr1 [612758, 612776]      + |     CCGGAGGCTGAGGTGGGAG          19
       [4]       chr1 [612758, 612776]      - |     CCGGAGGCTGAGGTGGGAG          19
       [5]       chr1 [629167, 629204]      + | ATTATTATAA...CAATCTTCCT       19T18
       ...        ...              ...    ... .                     ...         ...
  [872710] KI270751.1   [39996, 40045]      - | CAGGTCGCTG...AGAAAGGGAC          50
  [872711] KI270751.1   [64988, 65049]      + | TTTCTGACCT...AGACGTAAAC        55A6
  [872712] KI270751.1   [64988, 65049]      - | TTTCTGACCT...AGACGTAAAC        55A6
  [872713] KI270756.1   [45603, 45618]      + |        TCCGTTCCACTTTATT         9T6
  [872714] KI270756.1   [45603, 45618]      - |        TCCGTTCCACTTTATT         9T6
  -------
  seqinfo: 524 sequences from an unspecified genome; no seqlengths

I tried to realignment my fastq files using Tophat2, Bowtie2 nevetheless I have the same error.

Can you recommend the align tools and the exact parameters I need to use to get acceptable for wavClusteR bam files.

Thank you in advance,

Alexandr Gopanenko

Hi Alexandr,

thank you for your interest in wavClusteR. Can you please let me know about the exact nature of your fastq data? Is this a paired-end run? In principle, the choice of the aligner does not matter as long as gaps are not allowed and reads are single-end. wavClusteR has been thoroughly tested with bowtie, but I'm aware of some users who employed different aligners for pre-processing.

Hope this helps.

Federico