Nib files are supplanted by .2bit. The main reasons are: 2bit are twice as efficient as nibs in compression, and they incorporate maps for masking and N's that are space-efficient.
A .2bit file also has an indexed-directory in it which means that you can load it up with hundreds of thousands of sequences and it will not complain. A .nib file has no directory and can only hold a single sequence. This makes .nib a poor choice for assemblies with scaffolds, and since the .2bit can do everything that a .nib can do, it makes them obsolete. There are utils like faToTwoBit that will take care of the job. There ar twoBitMask bptForTwoBit faToTwoBit twoBitInfo twoBitToFa And many other libraries in our system know how to read .2bit files, including BLAT of course. twoBitInfo is handy as you can extract any sequence or any region of a sequence with it, or even just a list of all the sequence-names. -Galt Finney, Richard (NIH/NCI) [E] wrote: > There is a directory for hg18 nib files : > ftp://hgdownload.cse.ucsc.edu/gbdb/hg18/nib/ > But there is no similar directory in ftp://hgdownload.cse.ucsc.edu/gbdb/hg19/ > (causing great sadness). > > What's the thinking here? Are nib files evil? Obsolete? Coming soon? > What think you? > > > _______________________________________________ > Genome maillist - [email protected] > https://lists.soe.ucsc.edu/mailman/listinfo/genome _______________________________________________ Genome maillist - [email protected] https://lists.soe.ucsc.edu/mailman/listinfo/genome
