JASPAR2018

JASPAR2018 provides an open-access, curated, non-redundant database of transcription factor binding profiles represented as position frequency matrices (PFMs) and transcription factor flexible models (TFFMs) to support identification and analysis of TF-binding sites across six taxonomic groups.

Key Features:

• CORE collection expansion: The 2018 release adds 322 new PFMs (60 vertebrates, 262 plants) and updates 33 existing PFMs (24 vertebrates, 8 plants, 1 insect), representing a ~30% increase from the 2016 release.
• TFFM introduction: JASPAR 2018 includes 316 new TFFMs (95 vertebrates, 218 plants, 3 insects) to model flexible transcription factor binding preferences.
• PFM and TFFM representation: TF-binding profiles are stored as position frequency matrices (PFMs) and transcription factor flexible models (TFFMs).
• Clustering of PFMs: Clusters of similar PFMs are provided within each taxon and within each TF class per taxon.
• Human genome predictions: The CORE vertebrate collection was used to predict TF-binding sites genome-wide in the human genome, with predictions distributed via a UCSC Genome Browser track data hub.
• Programmatic access: All underlying data are retrievable via a RESTful API and the JASPAR 2018 R/Bioconductor package.
• Taxonomic coverage: Profiles cover species across six taxonomic groups, including vertebrates, plants, and insects.

Scientific Applications:

• Gene regulation studies: Models support analysis of transcription factor binding sites to investigate regulatory mechanisms governing gene expression.
• Comparative genomics: Cross-species PFMs and TFFMs enable identification of conserved regulatory elements across taxonomic groups.
• Predictive modeling: PFMs and TFFMs can be used to predict potential TF-binding sites within genomic sequences, including genome-wide predictions in human.
• Data integration: RESTful API and R/Bioconductor access allow incorporation of JASPAR 2018 profiles into bioinformatics analysis pipelines.

Methodology:

TF-binding profiles are represented as PFMs and TFFMs; similar PFMs are clustered by taxon and TF class; the CORE vertebrate collection was used for genome-wide TF-binding site prediction in human and distributed via a UCSC Genome Browser track data hub; data are accessible programmatically via a RESTful API and the JASPAR 2018 R/Bioconductor package.