NAUTICA: classifying transcription factor interactions by positional and protein-protein interaction information
- PDF / 1,129,200 Bytes
- 18 Pages / 595.276 x 790.866 pts Page_size
- 59 Downloads / 175 Views
RESEARCH
Open Access
NAUTICA: classifying transcription factor interactions by positional and proteinprotein interaction information Stefano Perna1* , Pietro Pinoli1 , Stefano Ceri1
and Limsoon Wong2
Abstract Background: Inferring the mechanisms that drive transcriptional regulation is of great interest to biologists. Generally, methods that predict physical interactions between transcription factors (TFs) based on positional information of their binding sites (e.g. chromatin immunoprecipitation followed by sequencing (ChIP-Seq) experiments) cannot distinguish between different kinds of interaction at the same binding spots, such as co-operation and competition. Results: In this work, we present the Network-Augmented Transcriptional Interaction and Coregulation Analyser (NAUTICA), which employs information from protein-protein interaction (PPI) networks to assign TF-TF interaction candidates to one of three classes: competition, co-operation and non-interactions. NAUTICA filters available PPI network edges and fits a prediction model based on the number of shared partners in the PPI network between two candidate interactors. Conclusions: NAUTICA improves on existing positional information-based TF-TF interaction prediction results, demonstrating how PPI information can improve the quality of TF interaction prediction. NAUTICA predictions - both co-operations and competitions - are supported by literature investigation, providing evidence on its capability of providing novel interactions of both kinds. Reviewers: This article was reviewed by Zoltán Hegedüs and Endre Barta. Keywords: Transcription factors, Interaction classification, Protein−protein interactions, TF-TF competition, Data-driven analysis
Background The classification of interactions between transcription factors (TFs) is foundational to the study of regulatory modules, i.e. groups of TFs implicated in the regulation of the same genes / transcriptional pathways. Classification based on localized binding-site information alone presents significant challenges, due to the confounding effect of intervening factors and the fact that some
* Correspondence: [email protected] 1 Dipartimento di Elettronica, Informazione e Bioingegneria (DEIB), Politecnico di Milano, Via Giuseppe Ponzio 34/5, 20133 Milan, Italy Full list of author information is available at the end of the article
interactions happen only in the regulatory regions specific to certain genes or in noncoding area. It is challenging to infer the precise nature of the interactions between two or more TFs, as they are dependent on their target, the cellular context in which the study is performed, and so on [1]. Transcription factors can compete to bind to a shared partner, compete for the same binding spots, or cooperate to coregulate some genes (and not others). Also, an investigation of all possible interactions between transcription factors (even for small genomes) is combinatorial in nature, and the cost of said wet lab experiments grows with the number of potential candidates.
© The
Data Loading...
