GTRD

Gene Transcription Regulation Database (GTRD) is a database of transcription factor binding sites identified from ChIP-seq experiments that were systematically collected and uniformly processed using special workflow (pipeline) for BioUML platform. Raw ChIP-seq data and experiment information were collected from:

• literature
• GEO
• SRA
• ENCODE.

Initial raw data were uniformly processed using specially developed workflow (pipeline) for BioUML platform:

• sequenced reads were aligned to reference genome using Bowtie2;
• peaks were identified using MACS, SISSR, GEM and PICS peak callers
• peaks computed for the same TF and peak calling method, but different experiment conditions (e.g., cell line, treatment, etc.) were joined into clusters
• clusters for the same TF revealed by different peak calling methods were joined into meta-clusters

Learn more about GTRD build process

GTRD database is freely available for non-commercial organizations.

Database statistics

GTRD uses 5072 ChIP-seq experiments for 476 human and 257 mouse TFs that correspond to 542 TFClass classes.

ChIP-seq experiments by species

Most of ChIP-seq experiments (61%) have corresponding control experiment.

Control experiments

General statistics:

Object type	Total count	Per ChIP-seq experiment
ChIP-seq reads	183.8 × 109	36.2 × 106
Reads aligned	146.9 × 109	28.9 × 106
ChIP-seq peaks	>100 × 106	depends on peak caller

In average each TF has been measured in 9.37 ChIP-seq experiments, 54% of TFs have been measured in more than one experiment.

The ten most studied transcription factors are listed bellow:

Transcription Factor	Number of ChIP-seq experiments
CTCF	282
AR	117
PU.1	103
ERα	92
c-Myc	79
C/EBPβ	74
NF-κB p65	70
GR	53
REST	51
GATA-1	51

Detailed statistics.

Database structure

The metadata concerning GTRD is stored in MySQL tables.

Each ChIP-seq experiment has a row in 'chip_experiments' table, which assigns id and stores basic information about experiment. 'chip_experiments' table has following structure:

Column	Description	Example value
id	Unique experiment identifier	EXP000489
antibody	Antibody used in chromatin immunoprecipitation	sc-345
tfClassId	Id in TFClass[1] database of target transcription factor, NULL for control experiments	6.2.1.0.1
cell_line	Studied cell line	HeLa S3
specie	Species latin name	Homo sapiens
treatment	Cell treatment or conditions	IFN gamma
control_id	Id of control experiment, NULL for control experiments or experiments without control	EXP000490

The links to external databases stored in 'external_refs' table:

Column	Description	Example values
id	Experiment identifier	EXP000489
external_db	External database name	GEO or PUBMED or ENCODE or SRA
external_db_id	Identifier in external database	GSM320736

GTRD uses following object identifiers:

Template	Object type	Example
EXPXXXXXX	ChIP-seq experiment	EXP000489
READSXXXXXX	Collection of ChIP-seq reads	READS000770
ALIGNSXXXXXX	Collection of read alignments	ALIGNS010001
PEAKSXXXXXX	Collection of ChIP-seq peaks	PEAKS010000

The relationship between these objects is provided by 'hub' table:

Column	Description	Example values
input	Input object identifier	READS000770
input_type	Type of input object	ReadsGTRDType
output	Output object identifier	EXP000489
output_type	Type of output object	ExperimentGTRDType

ChIP-seq reads, alignments and peaks links to experiments with hub table in the following way:

input	input_type	output	output_type
READS000770	ReadsGTRDType	EXP000489	ExperimentGTRDType
READS000771	ReadsGTRDType	EXP000489	ExperimentGTRDType
READS000772	ReadsGTRDType	EXP000489	ExperimentGTRDType
READS000773	ReadsGTRDType	EXP000489	ExperimentGTRDType
READS000774	ReadsGTRDType	EXP000489	ExperimentGTRDType
READS000775	ReadsGTRDType	EXP000489	ExperimentGTRDType
ALIGNS010001	AlignmentsGTRDType	EXP000489	ExperimentGTRDType
PEAKS010000	PeaksGTRDType	EXP000489	ExperimentGTRDType

Web interface

This page or section is a stub. Please add screenshots here!

Web interface to GTRD is available here. It provides capabilities for searching and browsing GTRD.

Start page

Start page contains search box and the links to browse all experiments in the repository tree, explore databases statistics and transcription factor classification tree.

Search capabilities

Search GTRD

Open ChIP-seq peaks

The ChIP-seq experiments contained in GTRD can be queried from the search box on the Start page. GTRD uses Lucene engine for indexing and quering ChIP-seq experiments that provides rich syntax for searching. The search can be performed by transcription factor(name or class), cell line, antibody or treatment/conditions.

For example, to search for STAT transcription factors enter stat* in the search field and press enter. You can restrict query to HeLa cells treated with interferon with following query:

 tfTitle:stat* AND cellLine:hela AND treatment:IFN

The list of matching ChIP-seq experiments will appear in the 'Search result' tab. Select one of them to view detailed ChIP-seq experiment information in the information box. The information box also provides links to the experiment data: reads, alignments and peaks.

To view peaks or alignments click the link in the ChIP-seq experiment information box, the track will be opened as table. The track can be exported by pressing 'Export' button or opened in the genome browser by pressing 'Open as track' button in the general control panel.

Repository structure

GTRD is organized in hierarchical Repository.

The GTRD/Data folder contains following items:

• experiments - ChIP-seq experiments metainformation
• sequences - Raw ChIP-seq reads in fastq.gz format
• alignments - ChIP-seq read alignments
• peaks - ChIP-seq peaks identified by MACS and SISSRs peak callers
• matrices - Position weight matrices for transcription factor binding sites
• site models - Models for recognition of transcription factor binding sites
• statistics - Summary GTRD statistics
• views - TFClass centric view of GTRD data avalable as tree-tables

GTRD/Dictionaries/classification is TFClass classification tree used by GTRD to reference transcription factors.