bigGenePred Track Format

The bigGenePred format stores annotation items that are a linked collection of exons, much as BED files indexed as bigBeds do, but bigGenePred has additional information about the coding frames and other gene specific information in eight additional fields.

bigGenePred files are created using the program bedToBigBed with a special AutoSQL file that defines the fields of the bigGenePred. The resulting bigBed files are in an indexed binary format. The main advantage of the bigBed files is that only portions of the files needed to display a particular region are transferred to UCSC. So for large data sets, bigBed is considerably faster than regular BED files. The bigBed file remains on your web-accessible server (http, https, or ftp), not on the UCSC server. Only the portion that is needed for the chromosomal position you are currently viewing is locally cached as a "sparse file".

bigGenePred File Definition

The following autoSQL definition is used for bigGenePred gene prediction files. This is the bigGenePred.as file defined by the -as= option when using bedToBigBed. Click this bed12+8 file for an example of bigGenePred input. In alternative-splicing situations, each transcript has its own row.

table bigGenePred
"bigGenePred gene models"
    (
    string chrom;       "Reference sequence chromosome or scaffold"
    uint   chromStart;  "Start position in chromosome" uint   chromEnd;    "End position in chromosome"
    string name;        "Name or ID of item, ideally both human readable and unique"
    uint score;         "Score (0-1000)"
    char[1] strand;     "+ or - for strand"
    uint thickStart;    "Start of where display should be thick (start codon)"
    uint thickEnd;      "End of where display should be thick (stop codon)"
    uint reserved;       "RGB value (use R,G,B string in input file)"
    int blockCount;     "Number of blocks"
    int[blockCount] blockSizes; "Comma separated list of block sizes"
    int[blockCount] chromStarts; "Start positions relative to chromStart"
    string name2;       "Alternative/human readable name"
    string cdsStartStat; "enum('none','unk','incmpl','cmpl')"
    string cdsEndStat;   "enum('none','unk','incmpl','cmpl')"
    int[blockCount] exonFrames; "Exon frame {0,1,2}, or -1 if no frame for exon"
    string type;        "Transcript type"
    string geneName;    "Primary identifier for gene"
    string geneName2;   "Alternative/human readable gene name"
    string geneType;    "Gene type"
    )  

Note that the bedToBigBed utility uses a substantial amount of memory: approximately 25% more RAM than the uncompressed BED input file.

To create a bigGenePred track, follow these steps:

  1. Create a bed12+8 bigGenePred format file that has the first twelve fields described by a normal BED file as described here. (You can also read about genePred here.)

    Your bigGenePred file must have the extra eight fields described in the AutoSQL file above: name2, cdsStartStat, cdsEndStat, exonFrames, type, geneName, geneName2, geneType.

    Your bigGenePred file must be sorted by chrom then chromStart. You can use the UNIX sort command to do this: sort -k1,1 -k2,2n unsorted.bed > input.bed
  2. Download the bedToBigBed program from the directory of binary utilities.
  3. Use the fetchChromSizes script from the same directory to create a chrom.sizes file for the UCSC database you are working with (e.g. hg38).

    Alternatively, you can download the chrom.sizes file for any assembly hosted at UCSC from our downloads page (click on "Full data set" for any assembly). For example, for the hg38 database, the hg38.chrom.sizes are located at http://hgdownload.soe.ucsc.edu/goldenPath/hg38/bigZips/hg38.chrom.sizes.
  4. Create the bigBed file from your sorted bigGenePred input file using the bedToBigBed utility:
    bedToBigBed -as=bigGenePred.as -type=bed12+8 bigGenePred.txt chrom.sizes myBigGenePred.bb
  5. Move the newly created bigBed file (myBigGenePred.bb) to an http, https, or ftp location.
  6. Construct a custom track using a single track line. Note that any of the track attributes listed here are applicable to tracks of type bigBed. The basic version of the "track" line will look something like this:
    track type=bigGenePred name="My Big GenePred" description="A Gene Set
      Built from Data from My Lab" bigDataUrl=http://myorg.edu/mylab/myBigGenePred.bb
  7. Paste this custom track line into the text box in the custom track management page.

The bedToBigBed program can also be run with several additional options. A full list o the available options can be seen by running bedToBigBed with no arguments to display the usage message.

Example #1

In this example, you will use an existing bigGenePred file to create a bigGenePred custom track. A bigGenePred file that contains data on the hg38 assembly has been placed on our http server. You can create a custom track using this bigGenePred file by constructing a "track" line that references this file:

track type=bigGenePred name="bigGenePred Example One" 
  description="A bigGenePred file" 
  bigDataUrl=http://genome.ucsc.edu/goldenPath/help/examples/bigGenePred.bb

Paste the above "track" line into the custom track management page for the human assembly hg38 (Dec. 2013), then press the submit button. Note that line breaks have been inserted into the above track line for readability; they must be removed for the example to work correctly in the browser.

Custom tracks can also be loaded via one URL line. The link below loads the same bigGenePred track, but includes parameters on the URL line (line break inserted for readability):

http://genome.ucsc.edu/cgi-bin/hgTracks?db=hg38&hgct_customText=track%20type=bigGenePred
%20name=Example%20bigDataUrl=http://genome.ucsc.edu/goldenPath/help/examples/bigGenePred.bb

With this example bigGenePred loaded, click into a gene from the track. Note that the details page has a "Links to sequence:" section that includes "Translated Protein", "Predicted mRNA", and "Genomic Sequence" links. Click the "Go to ... track controls" link. There change the "Color track by codons:" option from "OFF" too "genomic codons" and be sure "Display mode:" is "full" then click "Submit". Then zoom to a region where amino acids display, such as chr9:133,255,650-133,255,700 and see how bigGenePred allows the display of codons. Click back into the track controls page and click the box next to "Show codon numbering". Return to the browser to see amino acid numbering.

You can also add a parameter in the custom track line, baseColorDefault=genomicCodons, to set the display of codons:

browser position chr10:67,884,600-67,884,900
track type=bigGenePred baseColorDefault=genomicCodons name="bigGenePred Example Two" 
  description="A bigGenePred file" visibility=pack 
  bigDataUrl=http://genome.ucsc.edu/goldenPath/help/examples/bigGenePred.bb 

Paste the above into the hg38 custom track page (removing the line breaks) to see an example of bigGenePred amino acid display around the beginning of the gene SIRT1 on chromosome 10.

Example #2

In this example, you will create your own bigGenePred file from an existing bigGenePred input file.

Note the above description in Example #1 on how to view genomic codons, including numbering.

Sharing Your Data with Others

If you would like to share your bigGenePred data track with a colleague, learn how to create a URL by looking at Example 11 on this page.

Extracting Data from bigBed Format

Since the bigGenePred files are an extension of bigBed files, which are indexed binary files, they can be difficult to extract data from. We have developed the following programs, all of which are available from the directory of binary utilities.

As with all UCSC Genome Browser programs, simply type the program name at the command line with no parameters to see the usage statement.

Troubleshooting

If you encounter an error when you run the bedToBigBed program, it may be because your input bigGenePred file has data off the end of a chromosome. In this case, use the bedClip program here before the bedToBigBed program. It will remove the row(s) in your input BED file that are off the end of a chromosome.