February+2012

February 2012
We started this month with doing 16 different PCRs testing 16 primers that we will use for Real-Time PCR. I will use the following for the PCR reactions:
 * ~ Reagent ||~ Microliters ||
 * PCR mix || 12.5 ||
 * Forward Primer || 2 ||
 * Reverse Primer || 2 ||
 * cDNA || 2 ||
 * Water || 7.5 ||

I have to test all the primers to which ones will work best for Real-Time PCR. After making the mixes, they must denature at 95 Celsius for 2 minutes and then sit again at 95 for another 15 seconds. Then to anneal, they will incubate at 60 Celsius for 1 minute and then sit again at 72 for 1 minute. And by looking at the picture below, this was a picture of our DNA to make sure it was good to use for PCR and seeing the peak at 260 tells us it is with no contamination. And the concentration is 32.4 which is pretty good. the cDNA is made from mRNA for our reactions.

After looking at the PCR results on a gel, we can conclude which ones worked the best: The best results will be around 100 base pairs which we can see using the ladder on the gel along with the one above. The real time pcr products are all designed to have base pairs between 80-200. So by narrowing down primers using traditional PCR we can better estimate which will work best in real time. By looking at the gel, we saw that gene 1, gene 3, gene 8, gene 9, gene 12, gene 13, gene 14, and gene 16. You can also see that some PCR products have two distinct bands like gene 2 and gene 7 for example meaning there were other things in them and are not what we want to use for real time.

Now that we have the primers we would like to use, we can continue and now do real-time PCR in which I will analyze the graphs and calculate the standard error to put on the points. Quantitative PCR consists of 5 different time incraments; 12 h, 24 h, 48 h, 72 h, and 120 h. For each hour there are two different processes, wild type and morphant. In wild type, the normal gene expression will show and in morphant, the knockdown expression shows. Using those we can analyze the graph. In oder to do the graph, we need 3 replicates so that we can show standard error/deviation. To calculate the standard error is the standard deviation divided by the square root of n which in this case n=3. The x-axis on the graph is Deve Time points and the y-axis is log10 mRNA expression.



And this is an example of what the graph might look like ^

I also had to design the way the plate would look for our genes, each plate has 96 wells and we have to fill 180 so we used two plates. I figured the design based on the following: 5 time points x 2 for each time point- W-t and MO x 2 dilutions x 9 genes = 180 wells to fill. The plates are arranged from 1-12 horizontally and then a-h verically. Each gene itself takes up 20 total wells for the whole PCR process. I arranged 4 complete genes on a plate with 1 gene running half and half on the two. My design is shown below:

I also made 3 ampecillian plates that were used for the poster presentation on the same day as the Master Class. Two plates I just grew normal florescent colonies on and the other one I painted a picture on using the rest of the bacteria colony. At the forum, I learned about Annemarie's project in which she is testing the effects of caffeine on zebra-fish by using caffeine tablets. And for the Master Class, I heard a talk by Jamie, went and saw mini-particle colliders, saw a PowerPoint on nuclear physics and learned about the CMS detector and how to differentiate between W- and W+ and Z and whether each event was a muon or an election. I also learned more about what Francis's project is about. For example I learned about the different techniques he uses like sectioning and co-immunoprecipitation.

And after drawing my plane for the qPCR plates, we revised them and came up with a more efficient way of doing them that looks like the following: We changed it so that the primers went across horizontally and we did less primers. We filled all 8 rows and 10 of the 12 columns leaving the last two columns as a blank with nothing in it. And in order to fill in each well, I made a master mix for each primer including water, SYBR Green and the forward and reverse primers. After i inserted that into each well, I had to go back and insert individually each cDNA depending on whether it is morphant or wild type. I also had to remember that the SYBR Green is light sensitive so I kept it in a light sensitive tube on ice until the last minute and added it last to the mix. While pippetting everything, before I had to spin down all the reagents and mix them to make sure nothing re-clumped and to make sure everything was at the bottom of the tube before mixing. Then I was able to begin switching tips each time to make sure there was no contamination.
 * ~ Reagent ||~ Amount (microliters) ||~ Amplification ||~ Total Amount ( microliters) ||
 * SYBR Green || 10 || 12 || 120 ||
 * Water || 6.6 || 12 || 79.2 ||
 * F. Primer || 1.2 || 12 || 14.4 ||
 * R. Primer || 1.2 || 12 || 14.4 ||
 * ~ Total Amount ||~ 19 per well ||~  ||~ 228 master mix ||

Then after completing the plate, I was able to put it into the qPCR machine and set up the program that allows the process to happen. It takes about 2-4 hours to complete and before we had to centrifuge the entire plate. When Francis and I reviewed the results, they turned out relatively well with a few wells that had error. We looked at the different graphs but we mainly looked at one. It showed the amplification and there should be one relative peak for each row of similar cDNA. And that is how we found the error, so when we go back and use the data or retry it, we will have to use error bars to identify the mistakes or throw out the really bad data.

Below is a picture of how Francis taught me to put the cDNA into the wells :

Once again, the genes we know but are being kept private due to the data not being published yet.