[Previous] [Up] [Next]
Go backward to Physical Chemistry of DNA Binding  
Go up to Progress Report on Work to Date  
Go forward to Conclusion of Progress Report  

Experiments in Progress  

Although the simulator results are intriguing, their accuracy remains uncertain in a number of respects. It does not seem that further study of the DNA chemistry literature would enable us to improve our confidence in our model very much, because the sort of thorough scientific studies of DNA binding and primer extension that we would need to use simply have not been done yet. In this field, there is really, at present, no good substitute for doing actual experiments.

Therefore, we have begun some initial empirical investigations. Our preliminary goals for the moment are (1) to gain experience with the molecular biology laboratory procedures that will be required, and (2) to test the accuracy of our model's predictions for the affinities of oligos similar to ones we will use in our machines. To this end, we have designed an experiment in which we use 3 oligos having different numbers of mismatches (0, 2, or 4) against a certain site on a template, and test their ability to act as primers (that is, to anneal to the template and be extended by a polymerase) at different temperatures. This will give us an idea of the accuracy of our simulation.

For convenience, we are using as our template "pBlueScript II KS," a circular species of DNA, 2961 base-pairs long, which is commonly used by molecular biologists for inserting genes into bacteria. We chose for our mutation target a 24 base pair segment of the template containing several restriction sites (9) so that after running a CMM we can use these determine whether our mutations have occured.

For now, however, we are not running the complete CMM procedure, but rather just testing the inital stage of binding of mutagenic oligos and their extension by polymerase. Below are the sequence of the template site together with the three mutagenic oligos (X's showing locations of mismatches): 

     3'<- GGGGAGCTCCAGCTGCCATAGCTA -5'
 
          ::::::::::::::::::::::::
O1   5'-- CCCCTCGAGGTCGACGGTATCGAT ->3'
 
          :::X::::X:::::::::::::::
O2   5'-- CCCGTCGACGTCGACGGTATCGAT ->3'
 
          :::X::::X:::::::X::X::::
O3   5'-- CCCGTCGACGTCGACGCTAACGAT ->3'
The pattern of mismatches was chosen to be the same as that in our design for the 1-2 counter machine.

The figure below (see this figure) shows the simulator's predictions regarding the fraction of templates that will be bound to an oligo at equilibrium, for each of these three oligos with oligo concentration at 0.1 microMolar. We chose to test the validity of the predictions at three temperatures: 30 C, 47.5 C, and 60 C. At the lowest temperature, both O1 and O2 should bind. At the middle temperature, O2 should bind to about half the templates, and at the highest temperature, only O1 should bind.

  

Figure 6 : Theoretical prediction from simulator of the log base 10 of the odds of binding for the 3 oligos currently being studied.
The method we are attempting to use to test these predictions is to use each oligo as a primer for sequencing the template. Depending on how much of the oligo is bound, the bands on the sequencing gel should be darker or lighter. At the moment we are still attempting to refine our laboratory technique to the point where we can do this experiment without error. In the meantime, we may also try an alternate method of assessment wherein we run a couple of cycles of CMM and then use restriction enzymes to digest the products and determine how many of the new strands incorporated the desired mutations.

These experiments are of course only the beginning; hopefully they will soon be completed and then we will move on to the point of executing some of the simple machines which we have simulated.

Footnotes

 
9.
Restriction sites are special sequences whose presence can be easily detected.
- Michael P. Frank, September 12, 1995. Formatted using HyperLaTeX-1.3.

[Previous] [Up] [Next]