This page covers the basics of setting up your own hub:
As you build your hub, use the "Hub Development" tab on the Track Data Hub page to -check your hub for errors or to disable file caching to +check for errors or to disable file caching to see your changes immediately, rather than after the 300ms refresh rate.
-To begin, we want to provide an example hub.txt that has been created in a way +To begin, here is an example hub.txt that has been created in a way to make it easy to swap in your data URLs in place of our examples. It indicates what settings are required and includes many optional settings that can help elevate your tracks beyond the basics. Alongside these settings, it includes short explanations of how those settings work and how to configure -them, but there is also a version provided without these. +them. There is also a version provided without the explanations.
The first step in creating a track hub is to create your hub.txt file. Download the example hub.txt -and use this as a starting point, changing our default values to those for your hub. But, we'll also -provide the necessary settings here. These settings control how your hub is labeled in the interface +and use this as a starting point, changing our default values to those for your hub. +These settings control how your hub is labeled in the interface and contact information:
hub myExampleHub # a short, unique internal identifier for your hub, no spaces
# shortLabel and longLabel are how your hub is labeled in the Genome Browser interface
# shortLabels should be under 20 characters and longLabels under 70
shortLabel Example Hub
longLabel Example Hub for useOneFile option
useOneFile on
email genome-www@soe.ucsc.edu
genome hg38
If you have tracks across multiple assemblies, see the full track hub documentation.
-The most common track types are bigBed and bigWig, compressed, binary versions of -corresponding plain-text formats. Together they should cover much +The most common track types are bigBed and bigWig. Compressed, binary versions of +the corresponding plain-text formats. Together they should cover most of what you might want to display in the Genome Browser, from transcription peaks to -RNA-seq results. +RNA-seq signals.
You can use bigBed tracks to display discrete annotations, such as genes, transcription start sites, or conserved genomic elements. The bigBed format builds off the plain-text BED format and is thus flexible in terms of what fields are included. -Your file must start with a set of 12 standard fields (though not necessarily all of them), but +Your file must start with a minimum of 3, and up to 12, standard fields, but can also extend the format with any number of additional fields.
-Next, we'll discuss how to build a bigBed from a bed file.
bedToBigBed utility for your system type from
our download
server.
bedToBigBed to build your bigBed:
bedToBigBed -sort in.bed chrom.sizes myBigBed.bbbigDataUrl setting in your hub.txt
+ other services.
+ bigDataUrl setting in your hub.txt.
Once you have built your bigBed files, it is time to create a stanza in your hub.txt file for that track. Here is what the required settings discussed above might look like for a basic bigBed track:
track bigBedRequiredSettings
shortLabel bigBed Required Settings
longLabel A bigBed Example with Required Settings
visibility pack
type bigBed 12 +
bigDataUrl gtexCaviar.chr7_155799529-155812871.bb
The type line consists of three parts:
Here is a screenshot of what this basic bigBed track looks like displayed in the Genome Browser:
The bigBed format also offer a wide range of customization options for the -display, from decorators to highlights. Additionally, they offer extensive -filter controls, searching options, and mouseover configurations. Our +filter controls, +searching options, and mouseover configurations. Our trackDb documentation contains a full listing of settings available for the format.
Here is the bigBed configuration with some commonly used settings, including filtering and mouseover configuration.
track bigBedCommonSettings
shortLabel bigBed Common Settings
longLabel A bigBed Example with Commonly Used Settings
visibility pack
type bigBed 12 +
bigDataUrl gtexCaviar.chr7_155799529-155812871.bb
filterLabel.cpp CPP (Causal Posterior Probability)
filter.cpp 0
@@ -174,90 +177,91 @@
And here is what that track looks like in the Genome Browser:
These common settings added options to the track configuration pop-up:
bigWig Tracks
You can use bigWig to tracks to display continuous
annotations, such as RNA-seq expression, conservation scores, or other
-genome-wide scores. You can build a bigWig using one of two plain-text formats:
+genome signal scores. You can build a bigWig using one of two plain-text formats:
wiggle or
bedGraph.
Building a bigWig
-Next, we'll discuss how to build a bigWig from a wig or bedGraph file.
- Download the
wigToBigWig utility for your system type from
our download
server.
- Use this utility to build your bigWig:
wigToBigWig in.bedGraph chrom.sizes myBigWig.bw
- If your assembly is a UCSC-hosted assembly (e.g. hg38), chrom.sizes can
- be a URL (replace "genNom" with the assembly name (e.g. hg38)):
- http://hgdownload.soe.ucsc.edu/goldenPath/genNom/bigZips/genNom.chrom.sizes.
- If you're working with a GenArk assembly hub, then the chrom.sizes
+ be a URL (e.g. hg38):
+
+ https://hgdownload.soe.ucsc.edu/goldenPath/hg38/bigZips/hg38.chrom.sizes.
+ If you're working with a
+ GenArk assembly hub (e.g. GCF/GCA), then the chrom.sizes
file can be found under the "Data file downloads" section on the
assembly gateway page.
- Put your bigWig file alongside your hub.txt in a web-accessible location,
either through the 10GB of space we
make available to users or through one of several
- other services
-
- You will use the file name (e.g. "myBigWig.bw") with the
bigDataUrl setting
+ other services.
+ - You will use the file name (e.g. "myBigWig.bw") with the
bigDataUrl setting.
bigWig track hub configuration
The basic trackDb configuration for a bigWig track is similar to a bigBed track as
-all tracks required the same basic settings (track, shortLabel, longLabel, type, bigDataUrl).
+all tracks require the same basic settings (track, shortLabel, longLabel, type, bigDataUrl).
This is what the configuration for a bigWig track might look like (the example
hub.txt includes other useful settings):
track bigWigExample
shortLabel bigWig Example
longLabel A bigWig Example with Commonly Used Settings
visibility pack
type bigWig -20 10.003
bigDataUrl hg38.phyloP100way.chr7_155799529-155812871.bw
color 60,60,140
The type line consists of two parts:
- "bigWig" is the basic track type
- "-20 10.003" indicates the minimum and maximum of the data in the bigWig
Here is what this looks like visualized in the Genome Browser:
Grouping tracks
Next, we'll provide a basic overview of how to group your tracks using
composite tracks and super tracks. This will allow you to pull similar data
-together under a single track.
+together under a single container.
Composite Tracks
Composite tracks can hold multiple tracks of the same type. For example, you
use a composite to group together a set of RNA-seq experiments including
replicates.
Here's what the configuration might look like for a composite containing two
bigWig tracks. There are two key components of a composite: (1) the line
"compositeTrack on" in the parent track stanza, and (2) including "parent
compositeName" for each track that will be part of the composite.
track compositeExample
shortLabel Example Composite Track
longLabel Example composite track using bigWigs
@@ -367,31 +371,32 @@
Sharing your hub
Once you have a functional hub that you would like to share with others, you
can create links that you give to others in two ways.
The first option is to create a session link,
which requires a Genome Browser account.
Load your hub, configure the genome browser as you'd like (e.g. position and
data tracks), select "My Sessions" under "My Data", and use the option to save
the current settings as a session. You will then be provided with a URL that
you can share with others.
The other option is to create a URL to the Genome Browser that loads your hub
-on the assembly of interest. There are three URL parameters you will want to use:
+on the assembly of interest. There are three URL parameters
+you will want to use:
db - UCSC assembly name (e.g. hg38)
position - chromosome position to load
hubUrl - URL to your hub
You will then append these to a genome browser URL. For example, this url with load the example hub:
https://genome.ucsc.edu/cgi-bin/hgTracks?db=hg38&position=chr7:155799529-155812871&hubUrl=https://genome.ucsc.edu/goldenPath/help/examples/hubExamples/hubBasicSettings/hub.txt
If you feel that your hub would be of general use to the research community,
you can contact us about making it a public hub. Note that public hubs have to meet
more stringent requirements than the basics described here.
Check that your hub meets the public hub requirements