Limits...
A computational model of quantitative chromatin immunoprecipitation (ChIP) analysis.

Xie J, Crooke PS, McKinney BA, Soltman J, Brandt SJ - Cancer Inform (2008)

Bottom Line: We developed a computational model of quantitative ChIP analysis to elucidate the factors contributing to the method's resolution.The most important variables identified by the model were, in order of importance, the spacing of the PCR primers, the mean length of the chromatin fragments, and, unexpectedly, the type of fragment width distribution, with very small DNA fragments and smaller amplicons providing the best resolution of TF binding.One of the major predictions of the model was also validated experimentally.

View Article: PubMed Central - PubMed

Affiliation: Department of Medicine, Vanderbilt University, Nashville, Tennessee 37232, USA.

ABSTRACT
Chromatin immunoprecipitation (ChIP) analysis is widely used to identify the locations in genomes occupied by transcription factors (TFs). The approach involves chemical cross-linking of DNA with associated proteins, fragmentation of chromatin by sonication or enzymatic digestion, immunoprecipitation of the fragments containing the protein of interest, and then PCR or hybridization analysis to characterize and quantify the genomic sequences enriched. We developed a computational model of quantitative ChIP analysis to elucidate the factors contributing to the method's resolution. The most important variables identified by the model were, in order of importance, the spacing of the PCR primers, the mean length of the chromatin fragments, and, unexpectedly, the type of fragment width distribution, with very small DNA fragments and smaller amplicons providing the best resolution of TF binding. One of the major predictions of the model was also validated experimentally.

No MeSH data available.


Related in: MedlinePlus

Results of computational simulation of quantitative ChIP analysis: normal distributionTF binding, quantified according to a scale from 0–1, is plotted as a function of position along a 1.5 kb linear DNA molecule. A single TF binding site (marked by green line) was assigned to a position 530 bp from its 5′ end. (A) Distribution of binding was determined for amplicons of 50, 100, 150, 200, and 250 bp, mean DNA fragment sizes of 100, 150, 175, 250, and 300 bp, and σ = 10% of mean. (B) Distribution of binding was determined for a mean DNA fragment size of 150 bp, amplicon of 50 bp, and σ = 0%–70% of mean. (C) Distribution of binding was determined for DNA fragments of identical size (300 bp, σ = 0) and amplicons of the indicated lengths. A normal distribution of fragment widths was used for these simulations.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
getmorefigures.php?uid=PMC2367313&req=5

f2-cin-6-0137: Results of computational simulation of quantitative ChIP analysis: normal distributionTF binding, quantified according to a scale from 0–1, is plotted as a function of position along a 1.5 kb linear DNA molecule. A single TF binding site (marked by green line) was assigned to a position 530 bp from its 5′ end. (A) Distribution of binding was determined for amplicons of 50, 100, 150, 200, and 250 bp, mean DNA fragment sizes of 100, 150, 175, 250, and 300 bp, and σ = 10% of mean. (B) Distribution of binding was determined for a mean DNA fragment size of 150 bp, amplicon of 50 bp, and σ = 0%–70% of mean. (C) Distribution of binding was determined for DNA fragments of identical size (300 bp, σ = 0) and amplicons of the indicated lengths. A normal distribution of fragment widths was used for these simulations.

Mentions: Using a range of DNA fragment sizes and σ = 10%, the model reveals that the smaller the amplicon, the more closely the binding isotherm flanked the binding site. Unexpectedly, for certain fragment widths (e.g. 300 bp), resolution varied inconsistently with changes in amplicon size, and better resolution was obtained for this fragment width with amplicons of 100 and 250 bp than 50 and 150 bp (Fig. 2A). Overall, however, resolution increased and signal strength declined with decreasing fragment size. In contrast, resolution was not significantly affected by variation in fragment width, and little or no difference in the binding isotherm was observed over a range of values of σ (Fig. 2B). Finally, when all DNA fragments were the same length (i.e. σ = 0) and the primers closely spaced, the model produced a “mesa”-like binding curve (Fig. 2C), exactly as described by the simple model of Kadosh and Struhl (Kadosh and Struhl, 1998). Thus, the type of fragment width distribution can interact with the variables of mean DNA fragment length and amplicon size in affecting the shape of the binding isotherm. Nevertheless, the greatest resolution using a normal distribution of DNA fragment widths was achieved with the smallest sized DNA fragments (e.g. 100 bp) and amplicons (e.g. 50 bp), similar to the exponential distribution.


A computational model of quantitative chromatin immunoprecipitation (ChIP) analysis.

Xie J, Crooke PS, McKinney BA, Soltman J, Brandt SJ - Cancer Inform (2008)

Results of computational simulation of quantitative ChIP analysis: normal distributionTF binding, quantified according to a scale from 0–1, is plotted as a function of position along a 1.5 kb linear DNA molecule. A single TF binding site (marked by green line) was assigned to a position 530 bp from its 5′ end. (A) Distribution of binding was determined for amplicons of 50, 100, 150, 200, and 250 bp, mean DNA fragment sizes of 100, 150, 175, 250, and 300 bp, and σ = 10% of mean. (B) Distribution of binding was determined for a mean DNA fragment size of 150 bp, amplicon of 50 bp, and σ = 0%–70% of mean. (C) Distribution of binding was determined for DNA fragments of identical size (300 bp, σ = 0) and amplicons of the indicated lengths. A normal distribution of fragment widths was used for these simulations.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
Show All Figures
getmorefigures.php?uid=PMC2367313&req=5

f2-cin-6-0137: Results of computational simulation of quantitative ChIP analysis: normal distributionTF binding, quantified according to a scale from 0–1, is plotted as a function of position along a 1.5 kb linear DNA molecule. A single TF binding site (marked by green line) was assigned to a position 530 bp from its 5′ end. (A) Distribution of binding was determined for amplicons of 50, 100, 150, 200, and 250 bp, mean DNA fragment sizes of 100, 150, 175, 250, and 300 bp, and σ = 10% of mean. (B) Distribution of binding was determined for a mean DNA fragment size of 150 bp, amplicon of 50 bp, and σ = 0%–70% of mean. (C) Distribution of binding was determined for DNA fragments of identical size (300 bp, σ = 0) and amplicons of the indicated lengths. A normal distribution of fragment widths was used for these simulations.
Mentions: Using a range of DNA fragment sizes and σ = 10%, the model reveals that the smaller the amplicon, the more closely the binding isotherm flanked the binding site. Unexpectedly, for certain fragment widths (e.g. 300 bp), resolution varied inconsistently with changes in amplicon size, and better resolution was obtained for this fragment width with amplicons of 100 and 250 bp than 50 and 150 bp (Fig. 2A). Overall, however, resolution increased and signal strength declined with decreasing fragment size. In contrast, resolution was not significantly affected by variation in fragment width, and little or no difference in the binding isotherm was observed over a range of values of σ (Fig. 2B). Finally, when all DNA fragments were the same length (i.e. σ = 0) and the primers closely spaced, the model produced a “mesa”-like binding curve (Fig. 2C), exactly as described by the simple model of Kadosh and Struhl (Kadosh and Struhl, 1998). Thus, the type of fragment width distribution can interact with the variables of mean DNA fragment length and amplicon size in affecting the shape of the binding isotherm. Nevertheless, the greatest resolution using a normal distribution of DNA fragment widths was achieved with the smallest sized DNA fragments (e.g. 100 bp) and amplicons (e.g. 50 bp), similar to the exponential distribution.

Bottom Line: We developed a computational model of quantitative ChIP analysis to elucidate the factors contributing to the method's resolution.The most important variables identified by the model were, in order of importance, the spacing of the PCR primers, the mean length of the chromatin fragments, and, unexpectedly, the type of fragment width distribution, with very small DNA fragments and smaller amplicons providing the best resolution of TF binding.One of the major predictions of the model was also validated experimentally.

View Article: PubMed Central - PubMed

Affiliation: Department of Medicine, Vanderbilt University, Nashville, Tennessee 37232, USA.

ABSTRACT
Chromatin immunoprecipitation (ChIP) analysis is widely used to identify the locations in genomes occupied by transcription factors (TFs). The approach involves chemical cross-linking of DNA with associated proteins, fragmentation of chromatin by sonication or enzymatic digestion, immunoprecipitation of the fragments containing the protein of interest, and then PCR or hybridization analysis to characterize and quantify the genomic sequences enriched. We developed a computational model of quantitative ChIP analysis to elucidate the factors contributing to the method's resolution. The most important variables identified by the model were, in order of importance, the spacing of the PCR primers, the mean length of the chromatin fragments, and, unexpectedly, the type of fragment width distribution, with very small DNA fragments and smaller amplicons providing the best resolution of TF binding. One of the major predictions of the model was also validated experimentally.

No MeSH data available.


Related in: MedlinePlus