Here's a tale from the lab.
Today we had an interesting discussion. It started off with PBS and ended up on the topic of understanding the principles behind much of the protocols used in a lab.
It all started when a rotation student asked if there was a lab stock of PBS (phosphate buffered saline), a common buffer used in the lab. Another postdoc informed the student that there wasn't a lab stock and she did not have any at the moment. The student had to either get some from another lab member or make the solution from scratch. Five minutes latter he showed up with a small packet of powder, one of these "add water, stir, and presto your solution is made" things. These packets are quite expensive in relation to the cost of making the solution from reagents off the shelf. I can understand that adjusting the pH manually is time consuming, however when I was a gradstudent (not so long ago) making these solutions wasn't so hard. We had calculated the proper amount of dibasic to monbasic sodium phosphate, so that if you added the right ratio of the two components you would end up with PBS with the right pH. We also had TRIS tables (they're at the end of the Molecular Clonning, a book we would simply call Maniatis after Tom Maniatis, the famed scientist and "senior author" of the book) to make TRIS buffers at various pH levels.
We then discussed another event. About two weeks ago, I was helping another student synthesize RNA in vitro. After purifying the newly made transcripts over a column, we had to exchange buffers. I told her to precipitate the RNA, "just like you would precipitate DNA" and resolubilize it in our preferred buffer. She then asked me whether she should precipitate it with TCA (trichloroacetic acid - a reagent that precipitates protein). I thought to myself, she had never precipitated DNA, could this be? And why should she? The only time she's touched DNA is when she prepares it with a kit. Why would she know how to precipitate DNA?
So my point? I don't want to rant about kits or premade, ready to use PBS packs (I've already commented on kits and subsequently got attacked by a couple of kit lovers). My point is that kits are black boxes. The more we scientists use these kits, the less we understand how basic lab techniques actually work. And the less we know how lab techniques work, the less we are able to trouble shoot. Now kits do have their advantages. Although homemade preparations often yield better results, kits are more reproduceable and reliable. I remember on many occasions when the pH of a single solution was off and all of my DNA preps failed, now that never happens. But if we want to get mRNA into cells, knowing the intricacies of RNA precipitations including which salts to include and what buffers to resuspend the RNAs in can make a huge differences.
Here's another example. Some individuals in the department have tried on several occasions to microinject tissue culture nuclei. They purchase prepulled microinjection needles and use automated microinjectors .. but in the end just can't do it. I pull my own needles and write my own programs for our needle puller, I constructed my own microinjection setup (with no electronics components) I even use a syringe instead of a pico-spritzer that automates the pressure flow into the injection needle. Consequently I am able to microinject nuclei reliably month after month. The difference is the dept of understanding. I can adjust my needles, regulate pressure and adjust many parameters of the microinjection procedure. Those that automate the whole process simply don't have the same capacity to adapt. Why do you need to adapt? Well not every cell type is the same. Different microinjected substances require different needle etc. I also have plenty of little tricks. It's all this knowledge that allows me to reliably perform this technique. Am I just good at it? Not really. I've already trained three others and they are much more capable of microinjecting than the other people who used the automated setup.
My point is that it is fine to use kits or automated microinjectors or premade PBS packs ... yes, it's fine to take the easy way, but try to at least understand how your kits work. Such information can help you troubleshoot, or adapt, and in the end that is the difference between average and "being good at the bench".
- Log in to post comments
Excellent post, Alex, and I couldn't agree more. I started my postdoc when kits just first started coming on the scene. My postdoc advisor would let us order a kit if we could convince him that, after adding up our time, it would be less expensive to the lab. What that meant was that you would have to know how to make the buffer, reaction, etc. first using basic reagents, thereby knowing the basis of the kit. Usually we could convince him it was cheaper to buy the kit.
As a prof on examining committees, I'd always ask candidates how their particular kits worked. Most knew the answers, mostly because word got out that I'd ask about these things.
The funniest part of that post by far is clicking through to the comments by the kit lovers (I like kits, but I have been accused of "old schoolism" before). "You're no Jacob Monod". I love it!! No one broke it to him, I take it, that Jacob and Monod are two different people?
I'd have to look up a protocol to precipitate DNA. I've never done it and I don't see needing to in the future. If a step is critical to ones experiment than knowing the details is important. I can see that. I certainly notice when my people make a mistake that should have been obvious if they had fully understood the goal of the experiment. Still. I can't know everything and purifying DNA is one thing that I'll leave to Qiagen. It is expensive, but so is my time and my grey matter.
All that being said it is kind of outrageous to buy premade buffers that take 5 minutes to make if you have the stock solutions prepared. That is lazy. About as bad as people buying prepoured agarose submarine gels.
I'm no biochemist or anything related to soggy, smelly or otherwise unsavoury substances, but perhaps this is a bit of a parallel to implementing well-known algorithms yourself as opposed to using a library?
I know how to write a well-functioning quicksort or a heapsort. I naturally did as a student and I've done it a couple of times since, either on platforms with no fully developed libraries or because I needed some unusual wrinkle in the sorting that the standard library could not provide.
That said, I'm happy to use the library versions whenever possible. They will generally not be quite as efficient as a dedicated version written specifically for the task at hand, but they are well-documented, generally bug-free, and take care of a lot of corner cases that are just a pain to cover yourself every time. I'm not just trading efficiency for convenience, but for dependability. If I'm using the library version, I can be confident that any issues I have will at least not be connected to the sorting, and so can other people reading my code.
I don't know how far (or if at all) this is analoguous to lab testing, but it seems to me it's not altogether dissimilar.
Oh, my, my, my! Today's "kit scientists" confirm Lord Rutherford's arrogant statement "All science is either physics or stamp collecting" and Annie Coulter's "....biologists are barely even scientists anymore. They're classifiers, list-makers, like librarians with their Dewey decimal system. Except librarians don't claim the Dewey decimal system holds the Rosetta Stone to the universe." And they are definitely not economists, overnight shipping on a $5 envelope of powder costs $50. Polly
Polly Anna,
A couple of notes on your acidic quotes (my-oh-my, are you by chance an angry scientist?). Kits promote laziness but they are also time saving and dependable. Yes, biologists who over-rely on kits tend to be more detached from their work (and less adaptable and less able to troubleshoot), but just because someone uses a kit it does not mean that biologists are stamp collectors or librarians with their Dewey decimal system. Having said that, some do act like stampcollectors, but others are definitely not.
Abel,
Sorry about your comment not appearing, it was classified by Movabletype as spam ... I'm guessing it's because of the "less expensive" and "cheaper". Just goes to show you that when you talk about money, the system pays attention.
Alex, I also don't know if it's the Pharmboy moniker - I get categorized as spam when commenting on my own blog. The "pharm" thing might also trigger span filters like those selling ViaG*a.;ra.
It must be, this last comment was listed as spam as well despite the fact that you signed into typekey AND I listed you as a "trusted commenter".
Nice post (but fix the spelling).