Limits...
An iterative strategy combining biophysical criteria and duration hidden Markov models for structural predictions of Chlamydia trachomatis sigma66 promoters.

Mallios RR, Ojcius DM, Ardell DH - BMC Bioinformatics (2009)

Bottom Line: The resulting model predicts the final training set with a high degree of accuracy and provides insights into the structure of the promoter region.This strategy and resulting model support the conjecture that DNA biophysical properties, along with RNA polymerase sigma-factor/DNA binding collaboratively, contribute to a sequence's ability to promote transcription.The proposed methodology is ideal for organisms with few identified promoters and relatively small genomes.

View Article: PubMed Central - HTML - PubMed

Affiliation: School of Natural Sciences, University of California, Merced, CA 95344, USA. rmallios@fresno.ucsf.edu

ABSTRACT

Background: Promoter identification is a first step in the quest to explain gene regulation in bacteria. It has been demonstrated that the initiation of bacterial transcription depends upon the stability and topology of DNA in the promoter region as well as the binding affinity between the RNA polymerase sigma-factor and promoter. However, promoter prediction algorithms to date have not explicitly used an ensemble of these factors as predictors. In addition, most promoter models have been trained on data from Escherichia coli. Although it has been shown that transcriptional mechanisms are similar among various bacteria, it is quite possible that the differences between Escherichia coli and Chlamydia trachomatis are large enough to recommend an organism-specific modeling effort.

Results: Here we present an iterative stochastic model building procedure that combines such biophysical metrics as DNA stability, curvature, twist and stress-induced DNA duplex destabilization along with duration hidden Markov model parameters to model Chlamydia trachomatis sigma66 promoters from 29 experimentally verified sequences. Initially, iterative duration hidden Markov modeling of the training set sequences provides a scoring algorithm for Chlamydia trachomatis RNA polymerase sigma66/DNA binding. Subsequently, an iterative application of Stepwise Binary Logistic Regression selects multiple promoter predictors and deletes/replaces training set sequences to determine an optimal training set. The resulting model predicts the final training set with a high degree of accuracy and provides insights into the structure of the promoter region. Model based genome-wide predictions are provided so that optimal promoter candidates can be experimentally evaluated, and refined models developed. Co-predictions with three other algorithms are also supplied to enhance reliability.

Conclusion: This strategy and resulting model support the conjecture that DNA biophysical properties, along with RNA polymerase sigma-factor/DNA binding collaboratively, contribute to a sequence's ability to promote transcription. This work provides a baseline model that can evolve as new Chlamydia trachomatis sigma66 promoters are identified with assistance from the provided genome-wide predictions. The proposed methodology is ideal for organisms with few identified promoters and relatively small genomes.

Show MeSH

Related in: MedlinePlus

Flowchart of duration HMM iteration.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC2743672&req=5

Figure 1: Flowchart of duration HMM iteration.

Mentions: Sources of error that could lead to misclassification include (i) imprecise laboratory procedures in defining and identifying promoters (including false positive promoters), (ii) presence of more than one promoter population, (iii) failure to include relevant predictor variables, and (iv) random variation. To minimize the first two error sources, an iterative strategy was developed. Duration HMM iteration (Figure 1) addresses error source (i), while SBLR iteration (Figure 2) addresses source (ii).


An iterative strategy combining biophysical criteria and duration hidden Markov models for structural predictions of Chlamydia trachomatis sigma66 promoters.

Mallios RR, Ojcius DM, Ardell DH - BMC Bioinformatics (2009)

Flowchart of duration HMM iteration.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC2743672&req=5

Figure 1: Flowchart of duration HMM iteration.
Mentions: Sources of error that could lead to misclassification include (i) imprecise laboratory procedures in defining and identifying promoters (including false positive promoters), (ii) presence of more than one promoter population, (iii) failure to include relevant predictor variables, and (iv) random variation. To minimize the first two error sources, an iterative strategy was developed. Duration HMM iteration (Figure 1) addresses error source (i), while SBLR iteration (Figure 2) addresses source (ii).

Bottom Line: The resulting model predicts the final training set with a high degree of accuracy and provides insights into the structure of the promoter region.This strategy and resulting model support the conjecture that DNA biophysical properties, along with RNA polymerase sigma-factor/DNA binding collaboratively, contribute to a sequence's ability to promote transcription.The proposed methodology is ideal for organisms with few identified promoters and relatively small genomes.

View Article: PubMed Central - HTML - PubMed

Affiliation: School of Natural Sciences, University of California, Merced, CA 95344, USA. rmallios@fresno.ucsf.edu

ABSTRACT

Background: Promoter identification is a first step in the quest to explain gene regulation in bacteria. It has been demonstrated that the initiation of bacterial transcription depends upon the stability and topology of DNA in the promoter region as well as the binding affinity between the RNA polymerase sigma-factor and promoter. However, promoter prediction algorithms to date have not explicitly used an ensemble of these factors as predictors. In addition, most promoter models have been trained on data from Escherichia coli. Although it has been shown that transcriptional mechanisms are similar among various bacteria, it is quite possible that the differences between Escherichia coli and Chlamydia trachomatis are large enough to recommend an organism-specific modeling effort.

Results: Here we present an iterative stochastic model building procedure that combines such biophysical metrics as DNA stability, curvature, twist and stress-induced DNA duplex destabilization along with duration hidden Markov model parameters to model Chlamydia trachomatis sigma66 promoters from 29 experimentally verified sequences. Initially, iterative duration hidden Markov modeling of the training set sequences provides a scoring algorithm for Chlamydia trachomatis RNA polymerase sigma66/DNA binding. Subsequently, an iterative application of Stepwise Binary Logistic Regression selects multiple promoter predictors and deletes/replaces training set sequences to determine an optimal training set. The resulting model predicts the final training set with a high degree of accuracy and provides insights into the structure of the promoter region. Model based genome-wide predictions are provided so that optimal promoter candidates can be experimentally evaluated, and refined models developed. Co-predictions with three other algorithms are also supplied to enhance reliability.

Conclusion: This strategy and resulting model support the conjecture that DNA biophysical properties, along with RNA polymerase sigma-factor/DNA binding collaboratively, contribute to a sequence's ability to promote transcription. This work provides a baseline model that can evolve as new Chlamydia trachomatis sigma66 promoters are identified with assistance from the provided genome-wide predictions. The proposed methodology is ideal for organisms with few identified promoters and relatively small genomes.

Show MeSH
Related in: MedlinePlus