[Bioc-devel] R long vectors not supported yet on R 3.3.2

[Wolfgang Huber]

The human genome is arranged in 46 chromosomes. The longest is ~250 Mb 
(~2^28). While a Hilbert curve layout of a single chromosome tends to be 
informative, there is no obvious meaning in treating the complete human 
genome as a single 3 Gb linear sequence.

Wolfgang

On 10/03/2017 21:54, Wolfgang Huber wrote:
> Two replies:
>
> 1. Downsampling?
> In case you want to use the Hilbert curve for visualisation, please note
> that you will need a graphics device with resolution 65536 x 65536 to
> display it. Many people have smaller screens, so binning the genome
> (e.g. into bins of 10x10=100nt) could be a practical solution, and more
> efficient than computing some large intermediate thing that your
> graphics device will then downsample anyway.
>
> 2. Long vector
> In case you really need the big curve: I just had a look at the C code
> in the "HilbertVis" package, which anyway uses long ints, and it does
> not look difficult to modify the R wrapper so that it uses a long
> vector. I assume that Simon would welcome the patch.
>
> Wolfgang
>
>
>
>
>
> 9.3.17 08:44, Sohaib Ghani scripsit:
>> I am trying to simulate hilbertcurve (of Bioconductor package) of
>> level 16 in R. It takes about 4^16=4 Billion points. I want to
>> generate the hilbert curve of genome (size about 3 billion).
>>
>> But I am getting this error
>>
>> long vectors not supported yet: memory.c:1668
>>
>> I am using 64 bit version (R 3.3.2) so my guess is I can use vectors
>> of length > 2^31. Also, my RAM is about 350GB.
>>
>> The command I am using is
>>
>> itr=4^16
>> hc = HilbertCurve(1, itr, 16, mode = "pixel", title = "pixel
>> mode",start_from = "topleft")
>>
>> Even when I am reading the whole genome sometimes R is crashing in the
>> process.
>>
>> I have read the other similar questions on this topic but could not
>> find the solution. Please help me what should I use for this problem.
>>
>>
>> Thanks
>>
>>     [[alternative HTML version deleted]]
>>
>> _______________________________________________
>> Bioc-devel at r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/bioc-devel
>>
>
> _______________________________________________
> Bioc-devel at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/bioc-devel

-- 
Best wishes
Wolfgang

-------
Wolfgang Huber
Principal Investigator, EMBL Senior Scientist
European Molecular Biology Laboratory (EMBL)
Heidelberg, Germany

wolfgang.huber at embl.de
http://www.huber.embl.de