FILE protocol

The FILE protocol can be used to retrieve data that is stored in files residing on your local file system, as opposed to being obtained from external web servers. Because of its speed and the fact that it does not rely on external servers, this is the preferred way of retrieving data when working with large datasets.

The recommended way to use the FILE protocol is to link it to a single file in one of the three efficient binary data formats:

2bit (for DNA sequences)
BigBed (for Region Datasets)
BigWig (for Numeric Datasets)

The protocol also support whole-genome files in BED, GTF and Interactions format, but such files are generally not recommended to be used as file sources unless they are quite small (at most a few MB).

In addition, the protocol can be used in a second "segmented" mode where genome data in either FASTA, BED or WIG formats (and only these!) are split into smaller segment-files in order to access the middle of chromosomes more efficiently. In this case, the filepath setting should point to a directory rather than a single file, and this directory should contain subdirectories named after each chromosome. Each of these directories should contain files with names on the format "chrXXX_nnn.[fasta,wig,bed]", where "XXX" is the chromosome and "nnn" is an integer number denoting the start position of the segment in that file (0-indexed). The size of each segment file is a parameter of the data format.

If the segment size is 0, it means that the data is not segmented, but the whole dataset is contained in a single file (and the filepath should point to this file)
If the segment size is at least as big as the largest chromosome, it means that the data is segmented across multiple files, but each chromosome is contained in a single file
If the segment size is greater than 0 but smaller than the largest chromosome, it means that each chromosome is potentially split across multiple files

For instance, for a RepeatMasker region dataset where the segment size is set to 100,000,000, the human genome should be split across the following BED files. (Note that using a smaller segment size than this is recommended for more efficient access!) The filepath setting should then point to the top-level "repeatmasker" directory. Note that regions that span across the segment boundary (i.e. start in one segment and end in another) should be included in BOTH segments!

repeatmasker/chr1/chr1_0.bed
repeatmasker/chr1/chr1_100000000.bed
repeatmasker/chr1/chr1_200000000.bed
repeatmasker/chr2/chr2_0.bed
repeatmasker/chr2/chr2_100000000.bed
repeatmasker/chr2/chr2_200000000.bed
repeatmasker/chr3/chr3_0.bed
repeatmasker/chr3/chr3_100000000.bed
repeatmasker/chr4/chr4_0.bed
repeatmasker/chr4/chr4_100000000.bed
repeatmasker/chr5/chr5_0.bed
repeatmasker/chr5/chr5_100000000.bed
repeatmasker/chr6/chr6_0.bed
repeatmasker/chr6/chr6_100000000.bed
repeatmasker/chr7/chr7_0.bed
repeatmasker/chr7/chr7_100000000.bed
repeatmasker/chr8/chr8_0.bed
repeatmasker/chr8/chr8_100000000.bed
repeatmasker/chr9/chr9_0.bed
repeatmasker/chr9/chr9_100000000.bed
repeatmasker/chr10/chr10_0.bed
repeatmasker/chr10/chr10_100000000.bed
repeatmasker/chr11/chr11_0.bed
repeatmasker/chr11/chr11_100000000.bed
repeatmasker/chr12/chr12_0.bed
repeatmasker/chr12/chr12_100000000.bed
repeatmasker/chr13/chr13_0.bed
repeatmasker/chr13/chr13_100000000.bed
repeatmasker/chr14/chr14_0.bed
repeatmasker/chr14/chr14_100000000.bed
repeatmasker/chr15/chr15_0.bed
repeatmasker/chr15/chr15_100000000.bed
repeatmasker/chr16/chr16_0.bed
repeatmasker/chr17/chr17_0.bed
repeatmasker/chr18/chr18_0.bed
repeatmasker/chr19/chr19_0.bed
repeatmasker/chr20/chr20_0.bed
repeatmasker/chr21/chr21_0.bed
repeatmasker/chr22/chr22_0.bed
repeatmasker/chrX/chrX_0.bed
repeatmasker/chrX/chrX_100000000.bed
repeatmasker/chrY/chrY_0.bed