On Fri, Mar 30, 2012 at 12:34 AM, Hiram Clawson <[email protected]> wrote: > See also: http://genomewiki.ucsc.edu/index.php/Big_file_converters > > Are you sure your wiggle file is legitimate ? 16Gb compressed
I forgot. 16Gb is the same in the format of chrom<TAB>pos<TAB>value So the actual wig file should be smaller (the wig files are generated on the fly and are piped to wigToBigWig, so I'm not sure how they exactly are.) Almost all the mappable positions have value. I have tried fixedStep. But it still failed showing the neeMem error. > sounds like it is much too verbose. Even if it had a data value > for each base in the human genome, it should be less than 5 Gb > compressed. For example, here is a set of files that cover the > complete human genome and their total size is 4.6Gb: > > http://hgdownload.cse.ucsc.edu/goldenPath/hg19/phyloP46way/vertebrate/ This directory has tracks for each chrom. So for wigfile that covers almost every base in the genome (mm9), it is better to generate tracks for each chrom? BTW, my machine has 72GB memory with a swap of 100GB. -- Regards, Peng _______________________________________________ Genome maillist - [email protected] https://lists.soe.ucsc.edu/mailman/listinfo/genome
