[Bioc-devel] GenomicFeatures FeatureDB EST Table Error

[Dario Strbenac]

I e-mailed UCSC and they said the preferred way is to download by FTP. Which means more lines of code to parse the text file into columns, then split the exons and widths columns up to be able to make GRanges.

---- Original message ----
>Date: Sun, 6 May 2012 22:36:53 -0700
>From: "Tim Triche, Jr." <tim.triche at gmail.com>  
>Subject: Re: [Bioc-devel] GenomicFeatures FeatureDB EST Table Error  
>To: D.Strbenac at garvan.org.au
>Cc: bioc-devel at r-project.org
>
>   Actually, try downloading the same thing from the
>   Table Browser and see if there isn't something like
>   the following at the tail of the file:
>   843  chr11  33910774  33910775  rs4756078  0  +  CC 
>   C/G/T  genomic  single 
>   by-cluster,by-frequency,by-2hit-2allele,by-hapmap,by-1000genomes
>   0.361204  0.223906  intron  exact  1 
>   SingleClassTriA---------------------------------------------------------------------------
>   procedures have exceeded timeout: 1200 seconds,
>   function has ended.
>   ---------------------------------------------------------------------------
>   (this is from my attempted download of the
>   snp135common track, but it appears to be happening
>   to you as well)
>   It would appear that we're being throttled.
>   On Sun, May 6, 2012 at 10:00 PM, Dario Strbenac
>   <D.Strbenac at garvan.org.au> wrote:
>
>     Hi,
>
>     It seems some data has been added to the EST data
>     table in UCSC that GenomicFeatures cannot parse.
>
>     > ESTs <- makeFeatureDbFromUCSC(genome = "hg19",
>     track = "est", tablename = "all_est")
>     Download the all_est table ... Error in scan(file,
>     what, nmax, sep, dec, quote, skip, nlines,
>     na.strings,  :
>      line 4587019 did not have 22 elements
>
>     The relevant lines of the session information are
>     :
>
>     R version 2.15.0 (2012-03-30)
>     Platform: x86_64-unknown-linux-gnu (64-bit)
>
>     other attached packages:
>     [1] GenomicFeatures_1.8.1
>
>     - Dario.
>
>     _______________________________________________
>     Bioc-devel at r-project.org mailing list
>     https://stat.ethz.ch/mailman/listinfo/bioc-devel
>
>   --
>   A model is a lie that helps you see the truth.
>   Howard Skipper


--------------------------------------
Dario Strbenac
Research Assistant
Cancer Epigenetics
Garvan Institute of Medical Research
Darlinghurst NSW 2010
Australia