PROMOTER PREDICTIONS

Bioinformatic analysis of transcription regulation: a modeling approach

 Partecipanti

Mappa

 Word cloud

Esplora la "nuvola delle parole (Word Cloud) per avere un'idea di massima del progetto.

 Obiettivo del progetto (Objective)

'Transcription is both the first step and a major regulatory checkpoint in gene expression. Transcription start sites are locations in genome where RNA polymerase initiates transcription, while transcription binding sites are locations where transcription factors bind to regulate transcription. Knowledge of both transcription start sites, and transcription factor binding sites, is crucial for understanding transcription. However, methods for bioinformatic detection of these sites, which are mainly based on information theory, are typically characterized by low accuracy. Major underlying problems are: i) transcription initiation is a complex process that is characterized by both binding of RNA polymerase and opening of two DNA strands, ii) transcription factor binding sites usually have to be discovered/aligned within longer DNA fragments, which is often technically demanding and unreliable, iii) discovery of direct target genes of a transcription factor is complicated by random occurrence of binding sites that have high binding energy, but are not functional in regulating transcription.

The main goal of our proposal is to develop bioinformatic methods for accurate detection of transcription signals. To address the above problems, we will use biophysical modeling to i) Develop a novel method for transcription start site detection in bacteria, which is based on explicit calculation of transcription initiation rates and takes into account both RNA polymerase binding and opening of two DNA strands, ii) Develop a method for inferring transcription factor-DNA interaction parameters directly from DNA fragments selected through high-throughput in-vitro selection experiments, iii) Develop a method for detection of target genes of a transcription factor, which detects an overrepresentation of binding energy distribution upstream of genes. We expect that these methods will significantly improve accuracy of analyzing transcription regulation.'

Introduzione (Teaser)

The EU-funded PROMOTER PREDICTIONS project will unravel the transcription regulation process using bioinformatics and biophysical modelling. This will enhance promoter prediction and the modelling of bacterial gene circuits.

Descrizione progetto (Article)

Transcription enables the synthesis of functional gene products like proteins, and provides a major control point in this process. In bacteria, transcription regulation allows their adequate response to changes in environment and their accommodation to different stages in life cycle. Transcription is exhibited by the enzyme RNA polymerase (RNAP), which binds to promoters through a sigma factor and initiates transcription at transcription start sites (TSSs). Accurate knowledge of TTSs is the first necessary step in understanding transcription and its control. However, available bioinformatic methods for TSS detection are plagued by inaccuracy, due to a high number of false positive results.

The 'Bioinformatic analysis of transcription regulation: A modelling approach' (PROMOTER PREDICTIONS) project significantly reduced false positive results by about 50 %, through systematic analysis of sigma 70 promoter elements in Escherichia coli (E. coli) bacteria. To understand mutual relationships between the promoter elements the researchers used a novel mix-and-match model for promoter recognition and quantitated specificity of RNAP for sigma 70 promoter elements through the corresponding weight matrices. These weight matrices helped produce aligned promoter elements with adequate transcription activity, and accurately identified strong promoters in a newly sequenced bacteriophage genome.

To further resolve and understand the reasons behind the remaining false positive results, the researchers computationally analyzed kinetics of transcription initiation by RNAP in E. coli genome. It was found that a number of poised promoters exist in genome, which are DNA sequences that have a high binding affinity to RNAP DNA-binding domains, but fail to result in functional transcription, due to too slow opening of the two DNA strands. Poised promoters are likely to be falsely associated with functional TSS, therefore providing a major source of the false positive results.

To gain a better understanding of transcription regulatory circuits in bacteria- researchers modelled transcript processing of clustered regularly interspaced short palindromic repeats (CRISPR) within the CRISPR/Cas bacterial immune system. The transcript processing is a crucial step in control of expression of small RNA (crRNA) molecules that recognise invading viruses. Research results helped in the development of a novel synthetic gene circuit for large production of useful molecules from small substrate concentrations.

PROMOTER PREDICTIONS research findings were published in six peer-reviewed journals and presented at one regional and three international conferences. Researchers will work on combining accurate promoter alignment with promoter kinetic parameters' calculation to further reduce false positives. Work is ongoing to resolve problems with detection of binding sites of transcription factors (proteins that regulate transcription),in order to improve the predicted specificity of their promoter binding. Successful project outcomes will significantly improve our knowledge of transcription initiation. The technical skills and methodologies developed could also be adopted in other research areas such as infectious diseases.