Email:
(optional)
Project name:
(optional)
Data type Microarray RNA-Seq Chip-Seq
Run with miRNA Yes No
Run in expert mode

Please provide email address, choose appropriate options, add files and click "Start upload" button.

Add files...
 

ISMARA - The Integrated System for Motif Actitivity Response Analysis is an online tool that models genome-wide expression or ChIP-seq data, in terms of computationally predicted regulatory sites for transcription factors (TFs) and micro-RNAS (miRNAs).

The only input required for running ISMARA is either expression data (microarray CEL files or RNA-seq BED/BAM/SAM alignment files), or ChIP-seq data (BED/BAM/SAM alignment files), from a set of biological samples.

Using the given gene expression or chromatin state data across a set of samples, ISMARA identifies the key transcription factors and miRNAs driving the observed expression/chromatin state changes, and makes detailed predictions regarding their regulatory roles. Results include, for each regulatory motif, inferred activities across the input samples, predicted genome-wide targets, enriched pathways and functional classes of genes, and direct interactions between regulators. All results are presented within interactive and easily navigable HTML pages. The full results are also available for download.

More extensive documentation and sample input data are provided under the "Usage" link. Please also note ISMARA's "Terms of use"

Citation:
ISMARA: Automated modeling of genomic signals as a democracy of regulatory motifs
Piotr J. Balwierz, Mikhail Pachkov, Phil Arnold, Andreas J. Gruber, Mihaela Zavolan & Erik van Nimwegen Genome Research 2014

Intro

The only input required to run ISMARA is either a set of microarray CEL files (gene expression data) or a set of BED/BAM/SAM alignment files (RNA-seq or ChIP-seq data) that contain the genomic mappings of the sequencing reads. Example input files are presented here. ISMARA normalizes all input data, and then uses them to calculate a quantitative `signal' associated with each promoter in the genome. ISMARA then models the promoter signals across the samples in terms of computationally predicted transcription factor binding sites in the neighborhood of the promoter (and, when analyzing gene expression data, also miRNA binding sites in the 3' UTRs of associated transcripts). As a result, ISMARA predicts the genome-wide regulatory interactions that explain the provided input data. These results include quantification of the importance of each regulator in explaining the input data, the activity of each regulator across the input samples, genome-wide targets of each regulator, direct interactions between regulators, and so on. A more extensive description of the results is provided here.

How to run ISMARA

Please submit the input files (CEL, BED, BAM, SAM) to the form above. Files can be submitted one by one or compressed into an archive (zip, tar.gz). Please specify the data type and, for RNA-seq or ChIP-seq data, please specify the genome assembly used for the mapped reads. We advice all users to provide an email address. A notification email will be send when ISMARA has finished analyzing the data, and this email also contains the URL of the results. After this, press the "Start upload" button. You will then be redirected to a page which shows the current status of your job. If you chose not to provide an email address, please make sure to bookmark or save the URL of this page!

ISMARA accepts different file formats. CEL files for microarray data and BED/BAM/SAM files for RNA-Seq and Chip-Seq data, which should contain the genomic mappings of the sequencing reads (see the sample data below). If you have more than 5 files, it is most practical to put all files into archive (zip, tar.gz, tar.bz2) and submit this archive to ISMARA. Note, the archive should contain just CEL/BED/BAM/SAM, i.e. files no folders or subfolders.

An "Expert" mode is available for experienced users that want to run ISMARA with individually defined sets of genomic regions and their own regulatory site predictions. An explanation is available here.

Supported affymetrix chips

Currently nine human and eleven mouse Affymetrix Gene Chips™ are supported. In addition, two Saccharomyces cerevisiae chips are supported.

Human chips:

  • HG-U133A
  • HG-U133B
  • HG-U133+2
  • HG-U133A_2
  • HuGene-1_0-st-v1
  • HuGene-1_1-st-v1
  • HuGene-2_1-st
  • HT_HG-U133A
  • HT_HG-U133B
  • HT_HG-U133_Plus_PM

Mouse chips:

  • MOE430A
  • MOE430B
  • Mouse430_2
  • Mouse430A_2
  • MoGene-1_0-st-v1
  • MoGene-1_1-st-v1
  • HT_MG-430A
  • HT_MG-430B
  • MG_U74Av2
  • MG_U74Bv2
  • MG_U74Cv2

Yeast chips:

  • YG_S98
  • Yeast_2

ISMARA results

A description of the results that ISMARA provides is given here.

Example results:

Epi-Mara results:

Arnold, Schöler et al. "Modeling of epigenome dynamics during differentiation reveals REST as a mediator of Polycomb targeting in neuronal progenitors."

The developers of ISMARA give permission to you and your institution to use the ISMARA webserver for internal, research purposes, on the following conditions:

  1. The ISMARA webserver will be used by you and/or your institution solely for non-commercial purposes, except with express permission from the authors.
  2. You may provide us with feedback on the use of the ISMARA webserver in your research, and we are permitted to use any information you provide in making changes to ISMARA.
  3. Any risk associated with using the ISMARA webserver is with you and your institution.
  4. The ISMARA webserver should be cited in any publication(s) reporting on data obtained by using it as:
    ISMARA: Automated modeling of genomic signals as a democracy of regulatory motifs Piotr J. Balwierz, Mikhail Pachkov, Phil Arnold, Andreas J. Gruber, Mihaela Zavolan & Erik van Nimwegen Genome Research 2014

Commercial users should contact us for licensing arrangements.