Its always something

Yesterday when taking data I was fighting with DNA issues. First there was too much and then it all but practically vanished. Then today, I had absolutely no problem obtaining a good coverage of DNA, but the NucS was practically nonexistant. I was using half the amount of buffer to combat this issue and twice the volume of the protein and yet I am still not really satisfied with the number of interactions I could see while filming.

I just started the analysis of the first video from today, so I will soon see if enough data points came out of it. It could be that the very few proteins I do see are actually the only ones interacting and that for some strange reason, BSA is being very robust today in preventing the non-specific interactions. I checked the concentration of the NucS on a slide all by itself and it really did seem fine so I do not think thats the problem.

Swan Song

Yesterday I analyzed all of Friday's data and saw a decent portion of interactions but most of them were really short as in only one frame which is not good for us because it could very easily be a nonspecific interaction if we observe it for such a short time frame.

Today I also had bad luck but in the lab. Experiment after experiment after experiment was bad. I wasn't breaking slides or anything but first I couldnt see any DNA, then I saw so much DNA the screen was practically white, then I couldnt see DNA again. After 18 of these, I gave up and switched to just taking videos of photobleaching. Which came out...okay.
Its not helping that both of today's lunch options included fries. I cant believe its harder to eat healthy in Paris than it is at college in America.

Tomorrow's my last day working, so I'm already getting nostalgic for the pipetting ;-)

Week in Review

So Tuesday I worked on my presentation so theres little to report there except i got a lot of advice and changes from Rachid and then in the afternoon from Cedric as well. On Wednesday I presented it to both of them as well as Antigone, a post doc, and two other grad students. I got some good feedback from everyone and made more changes to my powerpoint/spiel from what they said and the questions they asked. The main thing I have to keep in mind is not rushing through the whole presentation. Then we all went and had a nice homemade delicious lunch at Antigone's. It was just course after course after course of GREAT food. Four hours later we were back at work and completely unmotivated after such a long break with so many laughs.
But I made the latest recommended changes to my presentation and Rachid worked on the analysis of his data.

Thursdaym I did some experiments but they didnt turn out that great so we deleted the results. Then that afternoon I went to the musee d'orsay with two other interns which was a lot of fun and very impressive. We all became huge fans of Monet.

Today i started the day off by revising my Paper for this summer and then began on some experiments. Bumped into and had lunch with Haviland and Ginny for a change which was a ton of fun!

Putting it all together

My Presentation is pretty close to being done. We're still trying to fit a model to the latest data but so far the numbers aren't being too cooperative. I'll most likely update the graphs with next week's data before giving the presentation to my fellow interns and their advisors.
I also worked on my paper today and barring any huge developments over the next week, it too is done. Rachid has already read and approved it but I am looking forward to Cedric's opinions on my presentation on Wednesday.
Then we can get back to experiments!

Ye Olde Blog

I'm working on my presentation today. I'm giving it on Wednesday to Rachid, Antigone, and Cedric. I was supposed to do some experiments this morning first with one of our newest buffers but everyone in the lab with the plasma cleaner apparantly took this day off and half of our lab did too including the one person from LOB who also has a key to the plasma cleaner lab. I did help Rachid out with an attempted alignment of another laser but it did not work out even after hours and he believes there must be something subtly wrong with the setup somewhere that doesn't allow us to become precise enough.

Yesterday was Bastille Day so everyone had the day off from work. Kara and I took the opportunity to go get a really nice lunch and then afterwards some gelato! Then in the evening I went with half the group to go see the fireworks.

Tuesday and Wednesday were the usual mix of experimenting with buffers and performing analysis on the resultant data. Trends have been identified and recorded but we still have plenty of experiments to accomplish.

Pipettes for the win!

So after a long week of catching up on analysis, I spent today purely in the lab performing experiments. Today luck was on my side. Of the 12 slides I cleaned, none broke or leaked, and only two didnt produce good videos! I have now got 5 more trials with the NucS in water, 4 trials of NucS in buffer with 100 NaClm and a whopping 12 videos from photobleaching the NucS in the same buffer!

I am starting the analysis already and maybe will get at least one done before going home for the day.

Presenting

So Cedric will not be here the last week of my internship so I will most likely be doing my presentation twice. Once at LOB for him and probably Rachid and Antigone as well and then another time for the other interns and the advisors back at the University of Michigan.

As a result, I am spending my compiling time today trying to brainstorm what and how I want to present instead of looking up possible career options as a physicist.

I am spending the morning analyzing data which I also spent all yesterday doing. Some of the trials come out great others come out with very very few legitimate points. It has given us a lot of different datasets to compare but we still need to do more experiments so we can have more points under each dataset. Then hopefully our results will become statistcally signficant.
In the meantime, at least I now fully understand stochastic processes.

Pulling out the Big Guns

So Rachid went to talk to Cedric about my badge problem and he came to the rescue. He pulled out the form he had been sent which should have been sent to them as well and now a few minutes later I am in proud posession of an ecole polytechnique badge. My picture is horrendous but no more sandwiches!

Lies.

No badge. Miscommunication somewhere. We went yesterday afternoon and they didnt have it but told us to come back in the morning, we just headed down and the guy said he hadnt recieved any paperwork for it. Looks like I'm stuck eating practically meatless sandwiches for the rest of my time in France.

Lets sing the Xenia's got a badge song!

My badge is finally ready!!! I just have to go to the office and get my picture taken and Ill be able to get in and out of the buiding whenever I want!The theme song for this fortuitious event would be to the tune of Amy's got a gun but a lot funnier.

Spent the first hour this mornign doing analysis and then Matlab lost the connection so all my work was basicaly thrown away and of course since it couldnt connect I lost my license and had to wait for someone else to sign off matlab before I could recommence the analysis. So Rachid and I switched places and he did analysis while I worked on experiments this morning. Out of 7 slides I took 3 videos. We are seeing much fewer protiens than we're used to which means we get fewer good points out of each experiment. We might change the concentration to help remedy this.

Now Matlab is working again so I'm redoing this mornings analysis which got cut short and then after I finish up the data from yesterday, I'll hopefully have time to look at today's data.

Tonight is Girl's night! Inspired by our lovely evening together in Switzerland where we had a nice dinner and couldn't stop laughing, we are going to try and repeat the experience at an Italian restaurant in Paris.

Experiments in the morning and analyzing in the afternoon. The experiments went well, or so we thought until we sat down and started to actually look at the analysis. we're getting so few points on each one. its a real problem.

I also looked at the data from yesterday and those results are coming out horrendous too. I dont have a single point from three of the trials.

We're still experimenting with different salt concentrations so we'll probably redo these 4-6 trials and then analyze them before making a decision on whether higher salt does affect the longevity of NucS in solution.

Today was split up between analyzing data in the morning and performing more experiments in the afternoon. Rachid did the reverse which gave me a lot of independence but also the availability of his expertise for both portions.

In performing the analysis of the photobleaching of the NucS from last Friday , I found that the amount of time they survived was significantly greater (which is good!) than what I had previously found with Monday's data. I printed out the graphs and have saved them as images so theres a half decent chance theyll end up in my final presentation.

In the afternoom I did experiments with the full surface treatment and a different buffer that Rachid had used with his in the morning so that we could compare results. I started out with three slides and treated them all exactly the same as I went through the steps, yet one came out covered in DNA and the other two had a much more sparse density. Rachid doesn't know why this happens but we've seen it time and time again and is one of the reasons why its so difficult to optimize the surface treatment since variables you dont even control also affect each trial.

No four leaf clovers

A seriously unlucky morning as far as the experiments were concerned.

While Rachid worked with the students again, I was trying to do another set of photobleaching trials. I broke two slides just trying to set up, an inauspicious start to be sure. Then every ten minutes I checked the first slide to see how the concentration of NucS on the surface was developing. Even after an hour they were still fairly sparse. Then the laser seemed to exit TIRF mode on its own while I was adjusting the position of the objective lens. When I thought I had it back to how it was supposed to be, I couldnt find anything on the slide. I checked my other two slides but they were also void of NucS spots.

With the second set of trials, I saw no good areas on one. But! with the second one i did take ten videos. Looking at them again back at my desk though I have to say that although the points are very clear, there are so few of them that I'm not sure its a strong analysis either.

Unfortunately matlab is once again giving me the licensing error, so I cant test that theory. Back to reading papers!

Navigo? navino!

Another almost purely analysis day. I made my way through ten trials of photobleaching NucS to find the various time intervals each seeable protien took to photobleach. Throwing those into a gigantic vector I used them to help create my experimental curve for the interaction and have ever since been testing different numbers with matlab code for our theoretical curve to make it mimic as much as possible the experimental one. Most of the different variations dont take very long to compile but when I put a very small number in it can lead to quarters of hours disappearing.

This morning I went to a seminar on laser safety which was in french but the slides were in english. I was surprised by how much I was able to follow with this combination. I tended to lose the thread only when someone asked a question either because they spoke too softly, quickly, or perhaps just from the lack of a visual aid.

This afternoon, Rachid and I are doing another set of photobleaching experiments with NucS. We are going to try a different method that he recently read about in a paper to test its effect.

l'analysis du data

So far today I have already looked at and analyzed two of the datasets from yesterday. Neither one had a single reliable interaction. It takes roughly forty minutes for matlab to perform the analysis after we select our points so I have been spending that time trying to outline the flow for a new matlab program that would make use of the quantum dots to determine the position of the protiens. This is a neccessary step to improve the accuracy of our results and i'm sure would greatly benefit Rachid if i can make it work.

In the meantime Rachid is performing some experiments with NucS, specifically testing the number of nonspecific interactions that still occur even with a surface that has been passivated. So there is no DNA on the surface only our blocking reagent and the streptavidin.

Update: have now completed seven different analyses within one day. The number of reliable interactions are much fewer than we would have liked and a lot of them have too short of a time interval to be sure they interacted at all.

Just another manic monday (and Tuesday)

So today I did six different trials of the full surface treatment but with different buffers at the end to see if it affected the number of nonspecific protien interactions. Also this morning Rachid and I met again with Cedric and Antigone to talk about the status of the project. It was really interesting because most of the questions that get asked end up leading to basic physics principles and their direct application. For example, Rachid wants to use a different buffer but it would have to be heated to work, so Cedric pointed out that there would be a thermal gradient between the objective lens and the surface. This would induce some convective movements in the thin layer of oil in between the lens and the surface. In order to combat this we would have to make the objective lens the same temperature. Experiments with these types of requirements are normally done inside of a box which is kept at the same temperature always but this poses a problem for us because in order to align the laser it would be necessary to be able to see where it points on the ceiling. Additionally our frequent moving things in and out of the box would also change the temperature.

Then afterwards, Rachid and I went with Antigone to test which wavelengths of light the laser she uses has. We had to use a prism to seperate out the different frequencies and could clearly see a color difference between the various shades of blue and green. Because the laser has also seen many hours of work, two of the weaker frequencies had already stopped working and one of these was the one Antigone wanted.

Yesterday I did some data analysis until my matlab stopped working again :( and also as always assisted Rachid in the Lab with a few trials.

Lasers and Quasers

actually this post probably won't talk about quasars. See http://en.wikipedia.org/wiki/Quasar instead.

So yesterday we took down the old laser and put up the new old honker of a laser. He works beautifully but is rather on the noisy side as I was forewarned. We aligned all the mirrors and lenses which was really interesting for me since I had never done it before. Then finally after three days of sitting on our hands we did some experiments!
The results didn't come out too snazzy though.

Today, Rachid was working with the students, this is his last class with them, so I began some trials on my own. First I redid the photobleaching experiment. I put NucS on a slide for two minutes, washed it out, put water, and then looked at it with the microscope to find out how long it takes them to photobleac for analysis purposes. For ten of the trials I kept the NucS in water and for another ten I put them in the buffer. Contrary to theory the NucS in water came through a lot better. More were stuck to the surface and were brighter. I redid the entire set of ten trials with buffer after Rachid saw my data but it didn't come through that much better and he said to delete all 20. That's Science I suppose.

Headed to Switzerland this evening to see my Uncle and his family!!!!! I cant describe how excited I am to see them again.

I can type!

I finally took 2 minutes to try and figure out how to change my keyboard settings so it pretends to be layed out in the American style. It was so easy to fix! Now i'm back to typing at full speed! It was severely handicapping me. The best part is that there's a little switch on the Start Menu bar so anytime Iwant to change back for whatever reason (most likely because I can't remember where a specific symbol is) I just toggle it.

Yesterday there was a RER train workers strike so I had the day off from work. Today I came in hoping against hope that the laser would be checked quickly and we could get to assembling the new one and screwing the mirrors into their new places on the breadboard. However, its now 2pm and the manufacturers still haven't got here. I am still vaguely optimistic that I will do something productive today though.

Initially, this morning I was going to work on the matlab analysis and Rachid reexplained briefly to me some of the steps I needed to take. The second time around everything made more sense and I wrote up my own little matlab-how-to-analyze-your-data guide in my notebook. BUT tragedy struck and the first time through the main routine it ran for almost an hour before we gave up on it. I restarted matlab ready to try again but l'ecole polytechnique has a matlab license that only allows so many people to have permission to run things at any time and in restarting my matlab I lost my place in the queue. So I've been retrying the same command every ten minutes since then hoping some one else will log off matlab and I can begin doing work again. No luck yet though.

Also Cedric and Rachid took a look at the lamp this morning that Cedric uses in his experiments and it was having trouble also! It made me wonder if maybe it wasn't either instrument but actually the power supply in the room or something odd like that. However, Cedric replaced a part on the lamp and now its fine so my momentary hope that nothing was actually wrong with the laser was dashed. At least we can use the lamp if we need to.

Heading to Switzerland this weekend to visit my uncle and his family! Can't wait!

Looks like someone's got a case of the Mondays

First piece of news this morning was from Cedric saying that the manufacturer had confirmed that they wouldnt come until wednesday, thus making us lose three days of experimentation which considering that i am only here for 9 weeks is a weighty sum.

Rachid decided we could do at least some rough tests using the UV lamp as our light source instead of the laser. We were also using a DNA that had been marked with a different fluorophore (cy3 instead of alexa) so we needed to change the filters that the light passes through so that only the wavelengths that it absorbs and emits at would pass through. Rachid spent 2 hours screwing and unscrewing different lenses , checking their wavelengths, and testing them with the new DNA. We couldn't see anything. The camera wasnt even working at first but after we rebooted and restarted the program, the DNA still didn't show up. It was clear that the light we saw passing through the lens was not as intense as it should have been but it didn't matter what combination of filters with appropriate ranges we tried, it was still impossible to focus on any plane.

Rachid even tried looking at it optically but had no success. So I have spent the afternoon reading some scientific papers instead. The first one was titled "Visualizing DNA Replication at the Single Molecule Level". It focused on describing the exact step by step instructions for two different methods used to observe DNA replicating itself. The first method bonded beads to the DNA that could then be directly observed with a CCD camera. Depending on how much the beads are calculated to move, the analysis can state how much the DNA has either lengthened or shortened. I am still unclear as to why the DNA would shrink or lengthen and what the correlation between either change would say about the way it replicated itself. The paper was more of a manual so it did not describe the theory behind this behavior or even what their conclusions were after performing the tests. The second method used TIRF! Instead of watching the beads, the growing length of the DNA is observed directly. In both cases flow chambers were used to prepare the surfaces with layers of Blocking Reagent and streptavidin. Their flow chambers consist of three layers, the glass slide, doublesided tape with a channel cut through it, and a slide made from Quartz. The advantage of this method of layer deposition seems to be better control of the uniformity of each layer, but I will have to ask Rachid to be certain.
In reading this paper I focused on where their slide preparation methods deviated from our own as well as where they were similar. Other than the use of flow chambers the other greatest discrepency was that they sonicated their slides to clean them while we use a plasma cleaner. In comparison to their protocol ours is much faster.

The second paper I read was entitled "Single molecule Enzymology" and was largely a discourse on the importance of single molecule studies. The authors put forward that most experiments can only observe the overall average result of an experiment are likely to ignore the small percentage of contradicting results. As a result of this, a lot of information is lost because although the property you are studying maybe something that ought to be uniform to the group it could also be a property of dynamic disorder ( one which fluctuates with time individually for each molecule). To illustrate their claim evidence from one experiment on the enzymatic turnovers of a single cholesterol oxidase molecule were described in depth.

Playing Catch up

Social Side of Work:

So this week there was an LOB-wide Barbeque, and it was awesome! I had the opportunity to meet with a ton of people and made friends with this Greek Professor who now works in the states. She named off all these "dives" I should go to for salsa dancing. I also got to taste a lot of homemade french cuisine since it was a potluck-ish thing and there was nothing I didn't like.

On the bad side of social things, we played our third soccer game this week for the semi-finals and lost 1-0. Which means we're out of the championships. Everyone took it really well though and is talking about next spring already.

Work Side of Work:

Rachid and I were doing experiments trying to do a layer of blocking reagent and then a layer of quantum dots with only 4 to 6 showing at any time. We were having a lot of trouble getting the ratios of each right, and I suggested we put the Quantum Dots on first and figure out what concentration or time we need to only see four-six and then put the blocking reagent to block off the protiens. Rachid LOVED the idea. He pointed out that it'd give us so much more control over the experiment and would be easier to be consistent. He kept saying, "Yes, Yes, that would be good, we're going to do that. Thank you!"

The laser broke. Thursday afternoon it would shut itself off every five minutes (generally when I was in the middle of a video, ugh). Then Friday morning it turned on twice and then after that wouldn't come back up. I showed Cedric and he called the manufacturer and they think the tube of Argon Gas has a problem. Its the most expensive part of the laser so they're not thinking about replacing it right now. Instead they're going to pull 0ut the slightly older laser they used to work with, which is in perfect condition. BUT makes two noises, that of a large Bee and a high pitched whistling...I'm bringing advil to work for the next week. On monday or wednesday the manufacturer is sending someone to look at it and make sure the problem is what we think it is. If they confirm our belief than Rachid and I are going to take apart the entire station and put up the other laser, realign all the mirrors, and hopefully start doing experiments by Friday. It'll be good work experience for me but Rachid's face when Cedric told him the news was pure anguish.

A New Desk! for the week.

So today I performed two different control experiments. In the first one I put quantum dots and DNA on a slide at the same time and waited five minutes before washing it out and adding buffer. I did four trials where in the first two trials I waited five minutes and in the second set of two trials I waited only 2 trials. The intention was to view whether it would be possible to distinguish the DNA from the quantum dots, since we are hoping to use the quantum dots as markers so that our measurements for the distances between proteins and DNA. It was not difficult to identify the quantum dots or to find the proper positioning for the microscope, but I am not positive I saw any DNA at all. The quantum dots are continuously blinking but all at different rates so it is difficult to determine where I see a particle that fades as it photo bleaches and where I just see a quantum dot becoming temporarily dark. I took three videos but the fourth slide (using the second protocol) leaked and had to be thrown out.

My guesses, without talking to Rachid first, for why the DNA was so difficult to spot would be
-The DNA got accidentally photobleached before being viewed by the microscope (it is almost two weeks old)
-The DNA is there but we have to look very closely to notice it next to the bright multitude of quantum dots
-The Quantum dots were put on the slide first so maybe it was more difficult for the DNA to find places on the slide to stick and thus were more likely to be washed out.

Then after lunch, I set up the second experiment. Here, I added Blocking Reagent with Biotin,waited ten minutes, washed it out, added streptavidin, waited ten minutes, washed it out, and viewed it on the microscope. Ideally since neither of these surfaces have fluorescent markers I should have seen nothing. Of course though, dust and dirt also glow some, although they photobleach even faster than DNA. After observing that it was generally dark, I added the protein and focused on the proper surface. This time the objective was to see how good the blocking reagent was. Ideally, the protein would never come too close to the surface because of the Blocking Reagent, or at least never stick there for several seconds. Instead I should have only observed general brownian motion with the proteins coming near the surface and then being "repelled" in the opposite direction. Without analyzing the video closely, it is difficult to measure the time frames in which the protiens appeared but it seemed that they were in general corresponding to theory.

I feel 'sciencey' today

So every Friday, Rachid teaches a lab class for high schoolers. i.e. is not available for me to pester with questions or to give me instructions. So we got here a little earlier. I ran off to quickly clean off the slides with the plasma cleaner. When I returned, we did a quick test with quantum dots to make sure everything was working. I found them quickly enough and everything looked fine, so Rachid gave me his ideas for the different trials I might do and left. We decided to test the effects that various incubation times of Streptavidin had on the concentration of DNA. So far we had always been doing ten minutes, but in some papers Rachid had read that a lot of people used Streptavidin with a concentration one tenth that of ours or incubated it for only 1-2 minutes.

First I set up my four slides in their holders and gave them all an equal treatment of blocking reagent; after ten minutes, i washed them out and put on some streptavidin. After five minutes i washed out two of them and gave them DNA. For the other two I waited five more minutes before washing out the streptavidin. The first one I looked at; I couldn't see anything. I was worried that it had leaked oto the microscope thus ruining the index of refraction. So i took it off and checked it for leaks. It was fine but the process rempved any chance of us using it for further experiments. The next one came out fine and I took a video. Then I took the first one of the five minute set. I searched probably for close to 15 minutes trying to see something on the slides andwould on occasion see some of the white dots but they were so faint that i assumed they were dust. I think maybe I accidentaly photobleached this slide at one point before putting on the microscope. Then I went to pick up the fourth one but noticed that the paper underneath it was wet, so since it was leaking I just cleaned it up and started preparing four new slides.

This time I kept Streptavidin on two of them for five minutes, and on the other two for only two minutes. I got one video from each type. One of the other ones leaked, and I didn't have time to do the last one.

I converted all the videos to Tiff, saved them to a USB, and went back to my desk to begin the analysis but imageJ shows only static at the beginning followed by white. Im trying to google the problem now but so far no luck.

Also there is an LOB barbeque on Tuesday!

They attack from three sides you know...

So today i became better friends with Lauren who is working on her phd, is from Spain, lives on cite universitaire, recognized me while we were waiting for the train, and currently has a hurt foot thus obliging me to step in and play soccer. I also bonded a little with another girl who's name i have forgotten and who asked me to come to lunch with her and some of the other young people so they could get to know me. Then when we realized the soccer game was at the same time as the lunch so instead she came for the second half to cheer us on along with another guy whos name may or may not be guy? pronounced gee i think.

The game was a ton of fun and my team won! Each time is required to have at least one girl playing all the time and if the girl makes a goal it counts for two points so they were very eager to pass me the ball and I played center forward. I wish my skills had been up to par, but at least I wasnt too unagressive and I did get to take one shot (which the goalie easily blocked). I also headed one ball and got the ball away from a player once or twice. I was blocking against a guy probably two feet taller for most of the match so he kept heading balls that i couldnt even jump and reach. Still it was a lot of fun and now we are still in the quarterfinals. Tomorrow I've been invited to play again so I must not have been too bad ;)

In terms of lab work..nothing really went very well. One of my solutions leaked onto the microscope lens which is really bad because then you might never get all the quantum dots off it. Rachid helped me clean it and he said he was fairly sure it would be fine. I have gotten a lot better at finding the particles with the microscope and tuning it only minutely. I have also gained confidence in pipetting things, asking people where things are, finding things on my own, and went and cleqned the slides this mornng with the plasma cleaner all by myself. I wear latex gloves almost the whole day and am still not used to the feeling.

Tomorrow our goal is to find the perfect method for putting just the right amount of quantum dots on a slide so we can use them to measure distances more accurately (for when the whole slide moves) without having them overpower our image. In an ideal world we would have one on each corner of our image when we film. So we can reference both the DNAs and Protiens positions to them, which would be more accurate.

Rachid also said if things keep gong well we might work on a simulation for all this but first we need more data.

p.s. Bonus points to anyone who knows the origin of today's blog's title.

That was unexpected

So there is a soccer tournament between all the labs at l'ecole polytechnique and they are required to have at least one girl on the team. the girl that usually plays cant tomorrow so i have been recruited. I hqvent played soccer competitively since middle school so this should be interesting...Still i am very excited to be included and i hope to make friends with some more people here :)

1 is the lonliest number

So today I am working on analyzing the seven different sets of data we took yesterday. I have finished the first two and am waiting for the third data set to finish compiling. Basically what I do is open up one of the videos we made of the dna/protien interactions and skim through it to find the clearest picture of the DNA. Then i use this matlab program called suivi that Rachid wrote. It takes the input parameters of the screenshot the dna is on as well as the total number of frames Then it opens up a picture of the frame with the dna andyou pick all the points you think are actually dna and not just dust. You can distinguish between the two because dna is brighter and lasts longer. So i click through the frames of the video and make sure that the points i am selecting last longer than 1 or 2 seconds. I normally pick around 30 points depending on the quality of the trial.

The program goes through each frame of the film,(generally around 2500) searching for high intensity points in the vicinity of the xy coordinates of the dna you picked. high intensit points hopefully correspond to proteins interacting with the dna but are often something else. Then the program generates a massive matrix with all the concievable bits of information you could want about every dna/protein interaction. It also makes a graph for each dna point you selected graphing intensity vs time.

I look through each of these graphs individually. On each two lines are plotted, one corresponding to a large area around the dna points, and another that is nestled closer and thus is generally mpre accurate. I search for spikes in these graphs which would indicate that something bright was right on top of my dna point. for each spike i check the xy coordinates of the bright spot and its distance from the initial dna. If the xy coordinates are rounded to a number ending with .5000 then the point is not accurate and should be thrown out. If the distance fro, the dna is greater than 1.2, the point is also thrown out because the error would be too large. For any points that do make the cut, I keep trackof the x,y coordinates,distance from, the dna, and the number of slides its visible for. This last number is the most important because the average lifetime will indicate what mechanism the protiens are most likely using to repair the dna. Right now we tend to average1-4 good interactions for every 30 dna we track for 2500 slides. This is obviously less than optimal and so we are going to try to improve this statistic by changing how we prepare the solutions (i.e. editing the incubation periods and wash outnumber of times)

Today I met Arnaud and Josephine who are also both interning here but for much longer than me. I went to lunch with the, and theyre a hilarious bunch. I spoke in french as much as i could and they alternated between english and french so i got to practice speaking and understanding french; oh happiest of days!

Science Prevails!

So today, Rachid let me do just about everything by myself! We did seven trials. Six of Protocol 1 and one of Protocol 2. In protocol 1, we wait ten minutes between adding each layer of Blocking Reagent, Streptavidin, and DNA. Each gets washed out once except the DNA which gets washed out three times. In protocol 2, we wait twenty minutes between each layer but the wash out times remain the same. We were supposed to have two trials with protocol 2 but the second one broke and we didn't have any slides left prepared and we have to walk all the way to another building to clean new ones.

For protocol 1 we saw great results and i got to practice finding the DNA with the microscope which is I think the trickiest part of my project so far. I also got to flash the DNA to hasten the photobleaching and then add the protein. I started out as a pretty bad protein-adder. Its important to not touch the liquid on the slide when adding the protein because then you move everything around and your whole analysis will be garbage. Instead you want to slowly drop the protein droplets onto the liquid dna solution.

It was great getting to do almost everything myself with Rachid always asking, What do we do next? or What did we forget? and getting most of the questions right.

On the bad side of things, Rachid said my ID card probably won't be available for another week! Which means I can't let myself in or OUT of buildings for another week. I always need someone with me. bleh. Its a bit of a problem. I can't even get to lunch and back on my own unless i just hang out by the door until someone opens it. stupid.

Tomorrow Rachid is not coming to work because he has to attend a seminar in Paris so I will be spending the day doing analysis on the six different trials we did today or as many of them as I can. and if by some miracle I do finish them all I think i'm going to take a look at the matlab codes that we use. Rachid said if i could optimize them or improve them or add anything to them then thats totally my perogative. Its too bad i dont have matlab on my laptop or i could look at them in the evenings too.

This past weekend we went to the Centre pompidou, Louvre, Notre dame, sat at a cafe with a coffee, and walked all over Paris. This next weekend we're going to Versailles and Marseilles.

When you hit the wall...climb over it!

So as promised, today I will write a list of some of the things that can go wrong and that have forced us to restart experiments so far. This will probably include some of the innate errors as well.

The biggest obstacle to everything is dust or just not keep things clean enough. In the mornings we go prepare 8-10 circular slides. We take them over to this entirely different building and rinse them with acetone and ethanol. Then we put them in a plasma cleaner. Before producing the plasma we create a vacuum in the chamber and after waiting a minute for all the air in it to disappear we turn the vacuum off and turn on the solenoids which produce a plasma that you can actually see! I will try to get a picture of it before the summer is over but basically it looks like a purple mist. The plasma is ionized and so all the particles on the slides are attracted and pulled off. This is the most effective method to clean the slides. Some other scientists use what is called a hypersonic bath. Basically this entails putting the slides in a solution that is then vibrated at a hypersonic frequency. The drawback to this method is it can only shake off particles that are larger than its wavelength. After the slides are done we put them in a special box Rachid made that we also clean with pressurized air. And then walk back to our building. Now its important whenever we use them to only touch them on the edges and only with tweezers, which can be frustrating when youre trying to align them properly. There is always some dust on them and this is visible and bright in the video although the lifetime of its brightness is significantly shorter than that of the DNA.

Contamination can also come from our pipettes if we are not careful to use a new tip for each new solution we use. Also my hands aren't the steadiest so whenever the tip of the pipette touches the outer rim of the test tubes instead of going straight in to the solution, I also have to change the pipette tip. The same is true for the tip of the mini vacuum we use to "wash out" our slides. Everything that comes in contact with something biological (the tips, the circular slides, regular slides, and our latex gloves) all get thrown away after being used. The only exception are the metal circle things we lock the circle slide in to. THese we rinse well with ethanol and then let dry. We only have four so we can only prepare four tests at one time.

Light is almost as much of an enemy as dust. The fluorescent molecules on the DNA and the protein have to be kept in the dark as much as possible to prevent their being photo-bleached. This means we keep them wrapped p in aluminum foil and tight in our hands when we are getting them out. We also work in the dark with just one side lamp.

Using or creating a bad concentration of something. If there arent enough DNA for instance? We spent the first day just playing around with concentrations.

Once we had a bubble of air in the oil on the lens so we had to clean off the whole lens with this special liquid and special lens cleaning paper. The lens is 10,000 euros-ish so its important to be very careful so that we dont scratch it.

And then of course you can always break something. Like a circular lens for instance which I did yesterday and Rachid did today. They're absurdly fragile.

Today was similar to yesterday except I got to do some more of the stuff on my own, so I think Rachid's trust in me is increasing!

Tomorrow...a fun blog about the plans for this holiday, four day weekend!

Science!

Today was very productive, in that I Learned A TON!

I worked with Rachid, mainly in a dark lab putting together slides or diluting solutions. But right at the end I got to do a tiny bit of analysis too.

So, first steps upon entering the lab. Turn on the water, the camera, and the switch for the laser. But keep the laser shuttered. Put the microscope on eye mode. Then prepare your DNA or Proteins. Generally when you make them, you make enough to last several days so often this step can be skipped. Today we made a DNA sample first by diluting DNA with water into 8 different containers. Afterwards we made the protein samples. We diluted the protein with a buffer that we also prepared. The buffer came from a mix of HEMES, NaCl, and water. One of each sample went into our ice bucket and the rest went into the freezer. Its important to keep each dark because the florescent molecules that are attached to the dna and the proteins have a two second lifetime under light.

Then we took some clean circular slides and locked them in between two metal circular objects. its important to center them first on one and to never touch the top side. We use a delicate pair of tweezers. Later in the day I tried to pick up one of these slides with the tweezers and squeezed to hard so it broke right there. I was more afraid of breaking them when screwing them in between the circular objects since it has to be tight so the substrate wont leak onto the microscope and change youre refraction angle and thus obscure your results.

Now in a different solution entirely there are quantum dots. So first put a small amount of these onto one of the slides you prepared (we could do up to five at a time) or onto a regular rectangular slide. Place it delicately on the microscope and lock it into place. Now switch the microscope to camera mode and make sure you notice it start on the computer. Now unshutter the laser. Initially it is probably going straight up and through the substrate, which means you are only performing epi-fluorescent microscopy. Move the platform that the final mirror that reflects the laser into the microscope is on. As you move it the light going through the substance changes angle until it is at 90 degrees and the critical angle is obtained. Now you are achieving Total Internal Relflection Fluorescent Microscopy!

The quantum dots are unaffected by the amount of light they have recieved so you should be able to fine tune your microscope until you see a thin layer of them blinking. Once you have acheived that you know that the microscope and camera are both working normally. Now test the Dna by using a pipette to drop a small amount onto the same coverslip but in a different area. move the slip so this drop is focused over the lens and make sure you can focus on the DNA and that there is an appropriate concentration.

Now you drip a little bit of DNA onto one of the circle slide, I think it was 50 nL? Let it sit for two inutes in the dark so that the DNA attach to the surface of the glass. Then wash it out three to four times with this special vacuum they have. Then place it into the groove over the lens. Before putting anything on top of the lens always! always! always! put a little bit of this special oil on to it first so that the refraction angle will be the same and to protect it. If you can find a good spot of dna on the camera focus in on it. Then turn the laser off, move the slide a bit to a new spot without dna that have already lost their fluorescence from the laser. Now start recording a video, turn on the laser, focus the microscope in on the DNA points, and wait for them to appear and fade. When all of them have faded you can move in to the next part. If there are too many of them you can flash the cover slip so that is fully dark again. Now drop an appropriate amount of the protein solution on to the slide. and watch as they glow. After most of them are gone stop the video, turn off the laser, and remove the slide. re-cover the microscope. Save the video to a .tif format. and voila!

Tomorrow: All the things that can go wrong!

Premier Jour!

So today we made it to the labs! I met Dr. Cedric Bouzigues and a graduate student, Rachid, with whom I'll be working all summer. My project is basically the current stage of his Thesis. Dr. Bouzigues took me on a quick tour of the facility, focusing on the room my desk is in and the lab the TIRF microscope is in. I worked with Rachid in the lab and already got to work with the microscope . I was trying to focus it on a line drawn on a cover slip first with 20X which was easy and then 100X which was trickier. Rachid says it will take some practice to get a feel for when to move closer or further away from the slide with the scope. Then as we discussed what was actually going to be done, I discovered that my project was quite a bit different from what I originally imagined which was probably part of the point of our whole presentations. However all the background research that I did was definitely helpful! I was able to answer a lot of questions correctly such as why we needed a liquid on the slides (so the refractive index is the same as the cover slip, the refractive index of the air is too different), knew that the slide was lighted from below by a laser, understood the overall objective of the project (to study how proteins repair broken dna), what gain is, and a couple other things I do not remember right now.

I will be spending my time this summer split between two general activities, taking data and analyzing it, for each stage. For taking data I will be preparing slides and then using the camera to take pictures and videos of the interactions between dna and proteins. I can not actually see the dna or the proteins but rather if either one is interacting they light up the entire pixel they are in. One of the sources of error in this project is the fact that both of these molecules are significantly smaller than the pixel itself so its not very well localized. Then when I analyze, I will be using matlab programs that Rachid has written to watch the pixels that seem to be dna and determine how many possible interactions each has per trial and how long these interactions last. When you first select the dna spot to study you determine its precise location with a matlab program. This is also within a margin of error because of the fact that the dna is smaller than a pixel but you obviously cant see smaller than a pixel and another error arises for this calculation because the slide is always moving a little bit from natural oscillations. One of the possible solutions to this problem would be to use quantum dots as markers for the slide and then measuring the relative positions of the dna and proteins to those quantum dots. Quantum dots are like the other molecules that have been dyed fluorescent but unlike those, their fluorescence doesn't ever die (the fluorescence of the dna and proteins disappears after an average of two seconds because of interactions with air molecules) because they are composed of inorganic materials.

To begin this summer my project is too optimize the surfaces to produce the most dna and protein interactions. To achieve this we are going to try covering the surface with a thin layer of proteins. The other proteins will not be able to interact with these proteins and so will hopefully be attracted more often to the broken dna "flaps". I am not yet sure how we cover a slide with proteins but I did get to watch Rachid prepare a slide today and place it on the microscope.

After we achieve this, I will be working with Rachid on studying the length of time interactions take to occur. The resulting average could help distinguish the strength of the current contradictory theories for the proteins behavior.

A whirlwind tour of Paris

One of the famous living statues of Paris

So yesterday after my favorite international flight ever, we arrived in Paris! After reaching the hotel our new ENSTA friends, Charles, Antoine, Neda, et Rania showed us all over Paris. We saw Le Tour Eiffel, L'Arc de Triomphe, Le Louvre, Notre Dame, et Le Champs-Elysses. But we didn't officially go into anything so there will still be plenty to explore in the coming weeks. Along the way we saw a lot of street performers including one of the Living Statues which was really cool also. I have never spent significant time in such a large city and am really enjoying the atmosphere. I am sure we will have a lot of adventures exploring the city further.

Tomorrow we are all meeting with our advisors and begin our main adventure. I am so ready to begin work on my project. I am sure at the beginning will be an adjustment period where I am asking a ton of questions and reading more papers in order to understand what exactly is going on. I am looking forward to perhaps a week from now when things have settled into more of a routine. Right now I am envisaging it being something like running or swimming in the morning, traveling to L'Ecole Polytechnique, working hard all day, traveling back, blogging and taking a small break, having dinner somewhere new with the other students or cooking something in the dorms, then studying for the Physics GRE for 1-2 hours, and then all hanging out or just relaxing, until we repeat it again. Weekends will probably be split between visits to other parts of France (a vineyard is high on my list!) or Europe, and exploring Paris in depth. I am sure the schedule will vary as we become absorbed in our projects and find we want to stay later some days to run a few more tests or something and other unprecedented events but I hope to study for the Physics GRE at least 4 times every week and work out in some fashion at least 3 times every week.

à tout à l'heure!

My Project Expecations Part Deux

Dna in an average human cell breaks 10 times per day. Natural proteins repair this damage and in general keep your genome intact. When mistakes are madem cell mutations can occur leading to serious problems such as cancer.

The various proteins that work to maintain your DNA in all of its proper coding are known but are not very well understood. Using TIRF microscopy this project is going to investigate the behavior of some of these proteins.

Figure 3: Mre11-2 and Mre11-1 repair broken DNA.

In the image above you can see the protein Mre11-1 acting to repair a strand of broken DNA. Mre11-1 is part of the MRN complex.

I will be working with the reparation protein NucS and observing its interactions with a free single strand of DNA. This is analogous to the types of DNA breaks that are seen in nature. In my case the protein will be fluorescently labeled so that we can see it clearly with TIRF. The goal will be to identify and study the mechanisms that NucS uses to repair the DNA.If this goes well we might move on to observing the interactions and cooperation among various proteins in the complex.

My Goals for the Summer:

-Learn how to prepare substrates

-Gain comfort and proficiency performing TIRF microscopy

-Increase independent thought in developing analysis
-Acquire a sense for graduate work abroad