?

Log in

Previous Entry | Next Entry

Life as a scientist really is a rollercoaster. I guess life in general is a rollercoaster. I hate rollercoasters.

When you've done a particular type of experiment many times, you become more efficient at it and obtain nicer results. Over the last week I did an experiment that I have done so many times that it should have gone very smoothly and given me nice data. It's a pleasant change from doing new types of experiments and screwing them up multiple times before you can get anything vaguely usable. But it's so much more frustrating when a reliable assay doesn't go well, for no apparent reason other than the universe hates me.

Here are potential things that can go wrong with this assay that should be resolved by now:
1. T cell prep. At the start, I was getting purities of 60-70%. You're supposed to get around 90%. Eventually we figured out this was due to an issue that wasn't my fault, and after fixing that issue and improving my technique, I can get 90% pretty consistently. This time, I got 98.4%. That's a record for me! And I had a good yield. I was happy.
2. Injections. Although I picked up the technique straight away when I first learned how to do injections, I've never been confident with them. I have good days and bad days, and I'm really slow. But I am improving, and this time I actually did pretty well with the injections. Almost all of them I hit on the first try, and the few where I had to try again, definitely managed to get the full dose. So I was happy with these too.
3. Reagents. I had to throw out almost a month's worth of experiments because nobody told me the reagents had to be kept frozen, rather than in the fridge. Even frozen reagents, I suspect lose activity after 2 freeze/thaws. But a lot of the stuff had 2 freeze/thaws already when it was given to me, and I can't just throw it out. Because I had such good purity from my T cell prep, I decided to use fresh tubes with no freeze/thaws, that should have full activity.
4. Mice. Sometimes we have to use whatever we have, and they can be of various ages and genders. We haven't seen signs that this has any major effects on the results, but it's less than ideal. This time, the mice were all of the same age and same gender. Another reason why this was going to be the perfect experiment.

When I went to get the results from this experiment, instead of the expected beautiful data, I see.. no cells. Where are my super pure T cells?! How can they just disappear?? Luckily, it seemed to be a problem with the machine, as I could see them perfectly fine on a different machine. Phew. Until I realised, I could only see the first 6 samples fine. For the rest, it looked like half had no cells. WHY. At this point, it was quite late, as I had wasted so much time trying to figure out why I couldn't see anything on the first machine. So I just ran all the samples and went home, refusing to think about it until the next day.

I've now finished analysing the data, and it's clear that it's completely useless. With half my samples written off, there's not enough left to compare anything to anything else. After thinking about it, I also think I know why it didn't work. And the reason is so stupid that I actually feel more depressed. Basically, to save time, I loaded 2 injections in one syringe. I've done this before when I've had to do a lot of injections, and I found that saving time actually improves the quality of the injections. The more time I spend in the injection room, concentrating, the more fatigued I get and the harder it is to do the injections properly. You also only have to load and keep track of half as many syringes. But that was with a liquid solution, not cells.

I think that the cells settled to the bottom of the syringe, so that the first mouse got nothing, and the second mouse got a double dose of cells. That would explain why exactly half the samples, in an alternating pattern, are useless. Why were there cells in all of the first six samples? Because they were injected first, and there wasn't enough time for the cells to settle.

I know it's better to know what went wrong so you can change it, but I'm so upset that what would have been a rare experiment where everything was perfect was ruined by such a trivial decision. At least now I know not to do that again. I also promised myself to cut down on consoling myself with food, because I am currently the heaviest I have been in my life, so all I can do is sit here and be depressed. Hey, isn't that what lj is for?

Latest Month

February 2015
S M T W T F S
1234567
891011121314
15161718192021
22232425262728

Tags

Powered by LiveJournal.com
Designed by Tomohito Koshikawa