Generating Random DNA Sequences

qpxu007
July 17, 2015

Purpose and Algorithm

DNA consists of four type of bases (A, T, C and G). Here, we will develop a R server for generating random DNA sequences. We will generate a single strand DNA of length n with a desired base composition (i.e. percentage of pA, pT, pC, pG are given by the user).

We first use a uniform random number generator (runif) to generate a vector r of length n, then, for each r[i] value, we assign A, T, C, G the following scheme:

alt text

We calculate the basic melting temperature Tm for each sequence.

Implementation and Usage

We developed and tested the functions (dnaseq, calTm, printDNA etc) in R first. The shiny UI app was then built on top these functions.

  • Generating DNA sequence
dna<-dnaseq(20)
printDNA(dna)

[1] "AGATCTAAGATTACCAGGCC"
  • Calculate Tm
print(calTm(dna))

[1] 49.73

Plotting Actual Base Distribution

plotDNA(dnaseq(200,pcg=90))

plot of chunk unnamed-chunk-4 

plotDNA(dnaseq(1000))

plot of chunk unnamed-chunk-5

Deployment to the shinyapps.io Web server

  • The application is deployed to shinyapps.io.

    • Slider bars are used for user inputs.
    • Inputs and results are displayed in real time.
    • The base distribution plot can be deactivated.

  • User can use it to generate random DNA sequences, and explore the Tm changes with regard to GC content and length.

  • Compare to other servers, e.g. knowpapa, ucr.

  • Formula for calculating Tm

  • Further reading