Andresen Lab Blog

This morning, I ran the second set of samples twice. The calibrations for the second run were a little funky but overall it mostly affected the wavelengths that we do not use in data analysis. This afternoon I have been organizing the four runs, averaging them together and creating graphs that match the ones for the 100x and 20x dilutions. 

Because we have only taken data for a high and low dilution, today I made and ran samples at a dilution between the previous samples. I created two sample sets along with new calibration solutions. This afternoon, I ran the first set of samples twice. Taking a quick glance at the data, I noticed that the concentrations for the different elements are not a steady increase or decrease as it was previously. Hopefully once I begin analyzing the data, these strange data points will not significantly affect the overall trends. 

Today I got to run all of my samples! First I ran 10 samples of 20x buffer solution twice to get smaller standard error. The error was reduced from about 0.14 to 0.04. I also ran a sixth run of the 20x samples for varied sodium by combining the left over solution from the other 4 runs. Unfortunately, this run measured sodium much lower than the other runs so I am not sure if we should include this data or not. 

Yesterday and today, I machined and fixed a rail that will be used to hold our XY stage. I also spent some time figuring out the step size for our motors. That is all.

I continued researching articles about the various parts of my experiment from ion competition to the different techniques used to investigate DNA compaction and aggregation including the ICP, tweezers, AFM, SAXS and different types of crystallography. This afternoon, the argon gas was replaced and the ICP began to warm up for the numerous runs that will hopefully take place the rest of the week. The first run will be with the 10 buffer solutions I just made at 20x dilution to decrease error bars. In the previous four runs at 20x, the concentration of sodium in the buffer solutions steadily decreased each run. Hopefully running ten more samples will even out the averaged concentration.

Today we had our first journal club meeting of the lab. Abby and I had read the paper Stretching DNA by Marko and Siggia. The paper introduces the theory behind the worm like chain model of polymers and applies it to DNA, which happens to be a semi-flexible polymer. The main importance of the paper is a useful formula which gives an approximate interpolation for the WLC force versus extension. This is a go to formula for most force extension work when it comes to force spectroscopy experiments with DNA. The main importance of this article however, is the fact that this interpolation formula is only useful for low and high end force ranges. At ranges in the middle, this formula is no longer accurate.

In addition to reading, I looked at more of the code today, trying to get the images displayed from the camera to the screen saved to file. What I found is that there exists a fatal error in the existing version of the code due to an inappropriate file call. What will happen if the user attempts to save the images to file is that the application will crash, and the user will be forced to stop the program through the task manager. This is of course not very useful.

Finally, Professor Andresen continued to teach me the basics of machining, introducing me to the procedures in operating the mill. We used the mill to machine a piece that attaches our XY table to the posts we machined earlier. This is what the lab looks like as of now:

Today I read and talked about reading, that is all.

Today I started by playing with the rotary motor. While it worked as expected following my port of the motor code to existing code that works for the other stepper motors, I had some difficulty figuring out what the exact unit was to make the motor do one full rotation. This actually took some time to calibrate. In the end, I had to find both how many steps the stepper motor took, and how many substeps the motor controller takes, multiply the two, and then for some reason multiply this number by three. This gives me how many sub steps are in one revolution for the motor. I then made this number be defaulted to 1, so that if the user entered in 1 for the unit the motor should turn, the motor will turn just 1 time. In order to ensure that the error in turning was minimized, I blew up the amount of turns the motor should take to 100. Thankfully, the motor stopped at the same spot it started at after 100 turns, indicating that the error is minimal.

By the afternoon, I had played with the code and camera interactions a little bit more. I noticed that there was a significant lag between what is displayed on the screen of the computer and movements that the camera actually sees. I believe, although I can’t test this yet since I don’t have the objective lens, that the portion of the code that generates the lookup table works as well.

The next steps will be to finish focusing the lens for the camera and to clean it. I also need to figure out how to attach the rotary motor, and the stepper motors to the XY table. There will probably be a significant amount of machining to do in the near future, which isn’t ideal, but necessary.

Today, I fixed the error bars on my graphs because I forgot to add my standard errors in quadrature. I also talked to Prof. Andresen to decide what to run next week to shrink the error bars. This afternoon I have been reading about optical emission spectrometers and I read the information for our round table tonight about research ethics. 

Today, I finished my PowerPoint and fixed all of the first run data of the cobalt- varied samples. I also read several journal articles including one about the Wormlike Chain Model that was very confusing. This afternoon, I showed my PowerPoint to Prof. Andresen and he noticed that the amount of sodium in the samples was significantly different between the 20x and 100x dilution. To try and correct this, I redid my analysis using all three wavelengths the ICP measured, averaged them all for each sample and added this to the figure to result in the figure below.

As you can see, the addition of the other wavelengths just shifted the data down but did not bring them closer to each other.