MeltDNA {DECIPHER} | R Documentation |
The denaturation of double-stranded DNA occurs over a range of temperatures. Beginning from a helical state, DNA will transition to a random-coil state as temperature is increased. MeltDNA
predicts the positional helicity, melt curve, or its negative derivate at different temperatures.
MeltDNA(myDNAStringSet, type = "derivative", temps = 50:100, ions = 0.2)
myDNAStringSet |
A |
type |
Character string indicating the type of results desired. This should be (an abbreviation of) one of |
temps |
Numeric vector of temperatures (in degrees Celsius). |
ions |
Numeric giving the molar sodium equivalent ionic concentration. Values must be at least 0.01M. |
When designing a high resolution melt (HRM) assay, it is useful to be able to predict the results before performing the experiment. Multi-state models of DNA melting can provide near-qualitative agreement with experimental DNA melt curves obtained with quantitative PCR (qPCR). MeltDNA
employs the algorithm of Tostesen et al. (2003) with an approximation for loop entropy that runs in nearly linear time and memory, which allows very long DNA sequences (up to 100,000 base pairs) to be analyzed.
Denaturation is a highly cooperative process whereby regions of double-stranded DNA tend to melt together. For short sequences (< 100 base pairs) there is typically a single transition from a helical to random-coil state. Longer sequences may exhibit more complex melting behavior with multiple peaks, as domains of the DNA melt at different temperatures. The melting curve represents the average fractional helicity (Theta) at each temperature, and can be used for genotyping with high resolution melt analysis.
MeltDNA
can return three type
s of results: positional helicity, melting curves, or the negative derivative of the melting curves. If type
is "position"
, then a list is returned with one component for each sequence in myDNAStringSet
. Each list component contains a matrix with the probability of helicity (Theta) at each temperature (rows) and every position in the sequence (columns).
If type
is "melt"
, then a matrix with the average Theta across the entire sequence is returned. This matrix has a row for each input temperature (temps
), and a column for each sequence in myDNAStringSet
. For example, the value in element [3, 4]
is the average helicity of the fourth input sequence at the third input temperature. If type
is "derivative"
then the values in the matrix are the derivative of the melt curve at each temperature.
MeltDNA
uses nearest neighbor parameters from SantaLucia (1998).
Erik Wright eswright@pitt.edu
SantaLucia, J. (1998). A unified view of polymer, dumbbell, and oligonucleotide DNA nearest-neighbor thermodynamics. Proceedings of the National Academy of Sciences, 95(4), 1460-1465.
Tostesen, E., et al. (2003). Speed-up of DNA melting algorithm with complete nearest neighbor properties. Biopolymers, 70(3), 364-376. doi:10.1002/bip.10495.
AmplifyDNA
, CalculateEfficiencyPCR
, DesignSignatures
fas <- system.file("extdata", "IDH2.fas", package="DECIPHER") dna <- readDNAStringSet(fas) # plot the melt curve for the two alleles temps <- seq(85, 100, 0.2) m <- MeltDNA(dna, type="melt", temps=temps, ions=0.1) matplot(temps, m, type="l", xlab="Temperature (\u00B0C)", ylab="Average Theta") legend("topright", names(dna), lty=seq_along(dna), col=seq_along(dna)) # plot the negative derivative curve for a subsequence of the two alleles temps <- seq(80, 95, 0.25) m <- MeltDNA(subseq(dna, 492, 542), type="derivative", temps=temps) matplot(temps, m, type="l", xlab="Temperature (\u00B0C)", ylab="-d(Theta)/dTemp") legend("topright", names(dna), lty=seq_along(dna), col=seq_along(dna)) # plot the positional helicity profile for the IDH2 allele temps <- seq(90.1, 90.5, 0.1) m <- MeltDNA(dna[1], type="position", temps=temps, ions=0.1) matplot(seq_len(dim(m[[1]])[2]), t(m[[1]]), type="l", xlab="Nucleotide Position", ylab="Theta") temps <- formatC(temps, digits=1, format="f") legend("topright", legend=paste(temps, "\u00B0C", sep=""), col=seq_along(temps), lty=seq_along(temps), bg="white")