Experiment 9: Analysis of Selection Results
Wednesday, July 12, 2000
Location: You will meet in room Z 744 for this week's experiment
(Verfügungsgebäude, 7th floor level).
Teaching Assistant: Your teaching assistant for this experiment is Charles
Information on the experiment.
The nuclease selection experiments performed last week led to a series of MALDI-TOF mass spectra (typically four per time point) that will have to be analyzed to identify compounds with increased survival time under nuclease attack. Supposedly, the compounds with increased survival time are in duplex form to the greatest extent, identifying them as high affinity binders.
The analysis can be performed on several levels. The first is a simple visual inspection of spectra at late time points, when the bulk of the competitors has been degraded and the "winners" of the assay can be identified directly as the most prominent peaks. It is recommended that you perform this form of analysis first and note the winner(s) before proceeding to the next level of analysis.
In order to obtain a representative survival score to be correlated with affinity, kinetics of the degradation of all compounds monitored have to be produced. In an ordinary enzymatically catalyzed reaction, one would try to get a rate constant from the initial rate of degradation. In this particular case, this type of analysis is problematic for several reasons. First of all, more than a single transformation of every compound occurs. The product of the first cleavage reaction (the n-1 oligomer) is again a substrate for nuclease cleavage, so that multiple substrates are being produced in the solution. Secondly, there are at least two phases in the degradation of a full length oligomer: the initial rapid degradation of unbound species, and the second, slower phase, which is probably off-rate limited (limited by the rate with which the strand is liberated from the duplex in the dissociation/association kinetics in solution). Thirdly, the number of data points available makes it difficult to do accurate fitting to get the initial rate of the reaction.
As a result, this laboratory has chosen to analyze the degradation kinetics by integrating the area under the individual kinetics of every component of a library. The assays are performed with a control compound (in your case the acetylated oligonucleotide), so that a relative area under every set of kinetics data points can be calculated. This relative area is termed "protection factor". A detailed discussion of this type of analysis can again be found in our recent publication on the nuclease selection technique (Altman et al., J. Combin. Chem. 1999, 1, 493-508.) Since it is not known how much the oligonucleotides will be protected from degradation prior to the assay, and we do not want to optimize the amount of enzyme or the reaction times individually (as this would be inefficient), a correction can be applied, if incomplete degradation of some library components occurred. The incomplete degradation would otherwise bias the protection factor analysis. We call this form of correction "cutoff and truncation". A schematic representation of how the correction works is given on the attached sheet, which is similar to Figure S1 in the Supporting Information of the above-mentioned paper.
The "experiment" itself will consist of a demonstration of a spectral analysis routine called "DAS", which is also available as a download on our homepage. After this, you are expected to analyze the spectra from your selection experiments, followed by a demo of a computer routine calles Automaton that uses the integration values from DAS. A few detailed comments follow below.
Analysis of spectra by DAS (demo). This program integrates
pre-selected regions in a series of spectra and normalizes the signal intensites
so that a desired peak height for the internal standard is obtained.
The data can be read into the analyzing program "Automaton", described
Analysis of your spectra for relative peak heights. The
computer routine DAS integrates regions of the spectra, but does not determine
peak heights. Even though the results are often close, peak integration
and peak height determination do not necessarily yield the same results.
In fact, the latter can sometimes be advantageous (see e.g. Sarracino &
Richert Bioorg. Med. Chem. Lett. 1996, 6, 2543-2548
and references cited therein). You are now asked to take the spectra
handed out by your teaching assistant and to graphically measure the peak
heights of the oligonucleotide hybrids and the internal standard in every
spectrum. For this, you will have to form groups of two, since not
all syntheses provided useful libraries.
Please take a pencil and draw a baseline through "the middle of the noise" in each spectrum. Now, measure the peak height from this baseline to the top of each peak of interest with a ruler. The numbers have to be compiled neatly on sheets (here it helps to have a partner who can measure while you are writing). The data should then be entered into a spreadsheet on a computer. Your teaching assistant will show you where this can be done. Please respect the privacy of the computers you are using and refrain from accessing files other than yours! From the raw peak heights, you have to calculate the relative peak heights (analyte/internal standard) for every spectrum. Of these, the mean and standard deviation are calculated for every time point and analyte. These are then normalized to the initial peak intensity for every compound, so that values that reflect the fraction of intact full length oligonucleotide at every given time point are generated (e.g. 0.35 with a standard deviation of ± 0.05). The data should be submitted to your teaching assistant who will enter them into a program that can do the integration of the kinetics and calculate protection factors. You can obtain the results from the integration from him on Friday.
Integration of kinetics with Automaton (demo). This computer
routine takes the kinetic data and performs the integration, together with
error analysis. One obtains the protection factors and, if required,
the protection factors corrected with cutoff and truncation. You
will get a demo from your teaching assistant and you will get results from
the analysis based on integration of peaks.
What you should also do for your final report. Besides
the details and results from your earlier experiments and the results of
the selection experiments in graphical form, your final report may also
contain other interesting details. You may want to analyze the spectra
from the selection experiments (and the syntheses) for side products, particularly,
if these side products showed prolonged survival in the selection assay.
You may also search for possible artifacts (e.g. partial degradation of
the internal standard). Finally, you can advance thoughts on how
a particular 5'-cap may interact with the duplex. Feel free to provide
an evaluation that goes beyond a list of numbers.