In a new in silico study providing an intriguing hypothesis for experimental verification, researchers in Iran have devised a computational approach based on HMMs (Hidden Markov Models) to determine how DNA topology regulates gene expression.
HMMs show the likelihood of a statistical relationship between variables and are widely used in bioinformatics and computational biology. A HMM describes a set of states corresponding to observed physical properties. For this study, the average free energies (AFEs) of the upstream DNA sequence of clustered (co-regulated) genes was used as said observed property.
The image above gives a schematic for the model, in which sequence X1, X2,…, Xn designate hidden regulating states and O1, O2,…, On the observed sequence or the calculated AFEs, which emit (E) from hidden states.

Regulatory sequences such as promoters not only contain cis-regulatory elements as switches of transcription, but also exhibit particular topological features…
The average free energy of promoter dinucleotides stacking nearest neighbors are subjected to scrutiny by statistical hidden Markov models to reveal the function of constrains and properties of promoter structure in transcription.
DNA curvature around origins of DNA replication regulate transcription through the RNA polymerase binding and formation of the open promoter complex, binding of transcription factors to their elements, nucleosome formation and chromatin remodeling.
…The proposed model monitors topological properties of each gene’s 50 upstream sequence structure and their related controlling states as input variables, and outputs the probability of having a set of co-expressed genes.

The validity of the theory requires the support that ‘‘genes with overlapping expression pattern during a particular time scale shall possess fairly similar structural properties in their regulatory regions’’.

In the framework of our claim, to acquire a proper relative conformation of DNA in each expression pattern the average free energy (AFE) or stability profile of the 5′ upstream 1000 nt sequence was calculated. The AFE is quantified directly from lengths of 20 nt sequence based on the nearest neighbor (NN) model. According to the NN model, a DNA duplex with different sequences could have almost identical AFE. Thus, it can be assumed that the regulatory sequence of co-expressed genes may have locally almost similar AFE. Consider the fact that the sequence structure is a precise result of base composition; the AFE can be a proper standard for assessing DNA structural features. However, because of experimental constraints in evaluating free energy values and restriction on base stacking composition in the NN model, the AFE can predict structural behavior of DNA oligomers accurately.

The authors’ model paints a picture of a regulatory network which is sequence structure to gene, not gene to gene. To this end, they were able to give quantitative predictions of transcript patterns simply by looking at stabilities of the gene regulatory sequence as depicted in the image above. The determinants of transcription levels in the model’s architecture are AFEs (or the topological profile of promoter sequence) rather than other transcript regulators.

It was revealed that DNA structural conformation plays a significant role in gene regulation, at mRNA level.
Secondly, DNA conformation, as a central key, has differential effects on the expression of various genes.
Thirdly, DNA conformation could be worthwhile to enlighten those parts of the transcription network which is imprecise to explain with other expression regulators.
Finally, the structural information which resides in 5′ upstream sequences determines the frame of the transcription network in the genome. Indeed, the relevant DNA sequence feature is the transcription network nucleoskeleton and strongly influences every protein’s attachment to the DNA surface…Thus as mentioned in previous studies, it could be assumed that the DNA nucleoskeleton conformation is a major factor which affects all aspects of transcription regulation.. 

Soltani et al (2013) The structural properties of DNA regulate gene expression. Molecular Biosystems, in press.

In a new in silico study providing an intriguing hypothesis for experimental verification, researchers in Iran have devised a computational approach based on HMMs (Hidden Markov Models) to determine how DNA topology regulates gene expression.

HMMs show the likelihood of a statistical relationship between variables and are widely used in bioinformatics and computational biology. A HMM describes a set of states corresponding to observed physical properties. For this study, the average free energies (AFEs) of the upstream DNA sequence of clustered (co-regulated) genes was used as said observed property.

The image above gives a schematic for the model, in which sequence X1, X2,…, Xn designate hidden regulating states and O1, O2,…, On the observed sequence or the calculated AFEs, which emit (E) from hidden states.

Regulatory sequences such as promoters not only contain cis-regulatory elements as switches of transcription, but also exhibit particular topological features…

The average free energy of promoter dinucleotides stacking nearest neighbors are subjected to scrutiny by statistical hidden Markov models to reveal the function of constrains and properties of promoter structure in transcription.

DNA curvature around origins of DNA replication regulate transcription through the RNA polymerase binding and formation of the open promoter complex, binding of transcription factors to their elements, nucleosome formation and chromatin remodeling.

…The proposed model monitors topological properties of each gene’s 50 upstream sequence structure and their related controlling states as input variables, and outputs the probability of having a set of co-expressed genes.

The validity of the theory requires the support that ‘‘genes with overlapping expression pattern during a particular time scale shall possess fairly similar structural properties in their regulatory regions’’.

In the framework of our claim, to acquire a proper relative conformation of DNA in each expression pattern the average free energy (AFE) or stability profile of the 5′ upstream 1000 nt sequence was calculated. The AFE is quantified directly from lengths of 20 nt sequence based on the nearest neighbor (NN) model. According to the NN model, a DNA duplex with different sequences could have almost identical AFE. Thus, it can be assumed that the regulatory sequence of co-expressed genes may have locally almost similar AFE. Consider the fact that the sequence structure is a precise result of base composition; the AFE can be a proper standard for assessing DNA structural features. However, because of experimental constraints in evaluating free energy values and restriction on base stacking composition in the NN model, the AFE can predict structural behavior of DNA oligomers accurately.

The authors’ model paints a picture of a regulatory network which is sequence structure to gene, not gene to gene. To this end, they were able to give quantitative predictions of transcript patterns simply by looking at stabilities of the gene regulatory sequence as depicted in the image above. The determinants of transcription levels in the model’s architecture are AFEs (or the topological profile of promoter sequence) rather than other transcript regulators.

It was revealed that DNA structural conformation plays a significant role in gene regulation, at mRNA level.

Secondly, DNA conformation, as a central key, has differential effects on the expression of various genes.

Thirdly, DNA conformation could be worthwhile to enlighten those parts of the transcription network which is imprecise to explain with other expression regulators.

Finally, the structural information which resides in 5′ upstream sequences determines the frame of the transcription network in the genome. Indeed, the relevant DNA sequence feature is the transcription network nucleoskeleton and strongly influences every protein’s attachment to the DNA surface…Thus as mentioned in previous studies, it could be assumed that the DNA nucleoskeleton conformation is a major factor which affects all aspects of transcription regulation.. 

Soltani et al (2013) The structural properties of DNA regulate gene expressionMolecular Biosystems, in press.

image/svg+xml
    blog comments powered by Disqus
  1. tfloscience reblogged this from biochemistries
  2. hotdogcephalopod reblogged this from molecularlifesciences
  3. fri624 reblogged this from molecularlifesciences
  4. amor-e-ishq reblogged this from molecularlifesciences
  5. vitabreva reblogged this from molecularlifesciences
  6. molecularlifesciences reblogged this from biochemistries
  7. biochemistries posted this