Tuesday, May 11, 2010

As described previously

Greetings Earthlings!

For this blog post, I urge you to close the PhD Comics/xkcd window that you have open and open http://www.nature.com. (Dude, if you don't know what comes up if you type this in the address bar, make sure your advisor doesn't know about it. (S)He'll have a fit.) There's a "Search" tool on the top right side of the page.

Click1 on the "Advanced search" option and type in the search term "as previously described" and from the drop-down menu2 menu, select "The exact phrase" and click1 on "Submit". If you have followed the instructions to a tee, this is what you should see:
->24,158 results for "as described previously"

Now do the same for "as previously described" as described previously2.
->22,531 results for "as previously described"

What this probably means is that there are a lot of half-sane people out there who are desperately looking for a protocol, but are not finding it where it should be! Why Nature even allows something like this is beyond me.

Until journals take a stand against incomplete and ambiguous protocols, be assured that you are going to play a never ending game of Hide (by the protocol writers) and Seek (which you will try with little luck).

Until next time,
Visual Acuity.

1 Take the cursor to it and press down down hard on the mouse at the place where your index finger is.
2 That thingy that has an arrow pointing down. When you click1 on it, it magically gives you a list of options.
3See how infuriating this is??

Tuesday, May 4, 2010

It's just human Nature

Greetings Earthlings!

In this post, we're taking it up a notch with a review of a protocol from a paper published in nothing less than the journal giant -- Nature! So put your hands together (not in applause, but in prayer; we're hoping to come out of this venture with a sound mind) for Ishiguro et al.'s 'Shugoshin–PP2A counteracts casein-kinase-1-dependent cleavage of Rec8 by separase' from Nature Cell Biology's April 2010 edition.

I'm sure it's a great paper, but let's quickly move to our section of interest -- "Methods". The Merriam-Webster online dictionary describes 'method' as "a systematic procedure, technique, or mode of inquiry employed by or proper to a particular discipline or art". Let's see if that definition holds true.

Take a look at the 'Chromatin Immunoprecipitation assay' (I have this sudden urge to binge on  potato chips and hence the ChIP protocol. All right, I agree that was a bad one.)

Chromatin immunoprecipitation assay.
1. Anti-rabbit IgG, anti-Rec8 [ref. 26] and anti-GFP (Living Colors Full-length A.v. Polyclonal Antibody; Clontech) polyclonal antibodies were used for immunoprecipitation.
2. DNA prepared from whole cell extracts or immunoprecipitated fractions was analysed by quantitative PCR with the ABI PRISM7000 system (Applied Biosystems) with SYBR Premix Ex Taq (Perfect Real Time) (TaKaRa).
3. The primers used for PCR were described previously [ref.25] except zfs1´ (5´-CAGTATTGCTGTACAAAGATGTTA-3´ (OLI559) and 5´-ATTTGCTAACTAGCCCAAAACGTC-3´ (OLI560)), 7E8 (5´-TGTCAACGTTGACAAGGAATCATT- 3´ (OLI561) and 5´-CCACGGAGTCAGCTACATCAATGT- 3´ (OLI562)), 1105(5´-GGTTTTGCCGATAACTTCTGTAAT-´ (OLI563) and 5´-CACCATTGGGAATTGCCAGCTTCA- 3´ (OLI564)) and abp1 (5´-CGTCCCCTGAGCGGAATTATTTTT-3´ (OLI565) and 5´- ATAAACTGGATTACCAGTAAGAGT- 3´ (OLI566)).
4. Immunoprecipitation by anti-rabbit IgG or by anti-GFP antibody in the untagged strain was accounted for as nonspecific binding in chromatin immunoprecipitation fractions of anti-Rec8 or anti-GFP, respectively.

FYI no one, I repeat, no one ever publishes protocols in papers with bullet points! This assay has been split up into readable chunks by yours truly for the sake of this blog post (Hey that's the anagram of blogspot! Please excuse my random rantings). That's another question for the protocol writers : isn't it hard enough  already to understand protocols? Why make them harder by writing them in the form of paragraphs?

Anyway, coming back to where we were, if you take a look at this protocol, you''ll see that there is none. In fact, it would be appropriate to rename it as Outline of
What-we-did-in-the-lab-to-get-the-results-we-got-and-aren't-in-a-million-years-letting-you-know-unless-you-are-prepared-to-grovel-and-ask. Let's attack the problem step-by-step.

Step 1 tells us what antibodies were used for immunoprecipitation. Great!

Step 2 makes DNA extraction and PCR seem like non-entities. Question time!
- How was the DNA extracted?
- What were the PCR conditions?

Step 3 again tells us how the primers used for the PCR (mentioned) in step 2 are the same as those used in a different paper's methods, with the exception of (the primers listed in the assay..). Incidentally, this step takes up the most space.

Step 4 tells us how to interpret "immunoprecipitation in untagged strain".

I just have one more question : how *does* one carry out this "Chromatin Immunoprecipitation assay"? I'm sure a lot of biologists have "Oh! The 'Methods' section is the least important of all the sections in the paper" or "It takes up too much space/time to write complete steps in a paper" or "Shania Twain is my favourite singer" as a response to this question.(OK probably not the last one)

For novices/robots all these reasons are immaterial. If you try to automate this protocol, you sure are going to have one confused robot.

Judgement time! So how many QM's does this protocol get? Being a democratic blog (except for the fact that I am the sole deciding factor in what goes into the blog), we will let the people decide again.

Until next time,
VisualAcuity

Tuesday, April 27, 2010

Pas si mal

Greetings earthlings!

In the previous post, we saw what it was that made a protocol make you feel more frustrated than a dog trying to catch its own tail. This time around, we will take a look at a protocol that is quite all right... until the very end. Like a Hitchcock movie, you know that something bad is lurking around the corner, but even when it hits you, you're surprised it was there. (If this doesn't make any sense to you, don't worry. It was not meant to.)

In the spotlight today is the Size selective DNA precipitation protocol from OpenWetWare.org. At first, it looks like a superior protocol - concise, complete and correct. Look closer, specifically, the last step.

"Dissolve the pellet in a appropriate amount of buffer of choice."

Questions:
1. Is 50 ml appropriate? Or a 1000 gallons? How much is "appropriate"?
2. What if my "buffer of choice" is blood plasma?(Blood plasma contains a bicarbonate - carbonic acid buffer)

[We have overlooked the previous step which is a paradox in itself:
"Carefully remove supernatant not to disturb the pellet, which will be invisible."
(It's like saying "Oh! Be careful with the dangerous pterodactyl that is right behind you! It's angry and invisible!! ")]

The BioCoder version marks this step as a to_do() instruction that will hopefully attract the attention of the person carrying out the protocol with its cheery red color (not to be confused with the red blood cells left over after extracting the blood plasma that we would like to use as our "buffer of choice".)

Also, the place holder for the "buffer of choice" is a generic buffer whose name and properties will have to be changed by the person executing the protocol. (In this case:
Fluid buffer = new_fluid("blood plasma", "bicarbonate + carbonic acid");)

Our final verdict:
2 QMs for this protocol because of ambiguity in two instances in the final step.
(And although we haven't taken the penultimate step into account, if we did evaluate it, it would get 5 QMs for playing with the sentiments of the poor guy/gal carrying out the protocol.)

Until next time, 
VisualAcuity.

Friday, April 23, 2010

Numero Uno

Greetings earthlings!

For our big début we have something truly special - something that gave could give you a minor concussion. Imagine this:

Your advisor wants you to come up with the protocol for perfecting his/her Potion for Making Grad Students Work Like Donkeys. So (s)he starts off by giving you some of her old Potion for Making Grad Students Work Like Donkeys (but which has only enough potency to make you work like a dog and that is the reason why (s)he needs a new protocol) and you dutifully start looking for the protocol. You are convinced that the protocol is in the paper by Kortylewski, M. et al. [1], specifically in the EMSA section.

Since you'd rather read a paper as a PDF, you download the PDF version of the paper first. You slowly make your way through all the text and get to the methods section, only to find that the materials and methods are only available in the Supplementary Information!

Ah, well, you know what they -- "No pain, no gain". So brushing aside this little slap on the cheek, you continue to fish out the methods in the Supplementary Information. You get there only to be greeted with this "EMSA and western blot analysis to detect Stat3 DNA-binding and protein expression were performed as described previously."

(Slaps Forehead). OK, let's get on with this. You then go to the paper this protocol can (hopefully) be found in. And if you thought it was as easy as this, I suggest you read your Grad Student Handbook ASAP. (Remember, Rule 1 : The student shall work like a donkey to get his advisor's work done.)

Sure enough, you do not get the protocol in the second paper[2] either. The second paper reminds you of a dentist who says, "I am just going to drill into your teeth a little. This will only cause a minor discomfort". The paper says "EMSA and western blot analyses to detect Stat-3 DNA-binding property and protein abundance, respectively, were done as previously described."

Oh no, the pain doesn't end here. You have to traverse through two more papers[3, 4], before you actually get the protocol[5]. Of course, by this time, you have lost a sufficient quantity of keratin-based substance on your head (no wonder so many Grad students have an especially bad case of alopecia).

At the end of your quest, you have a minor concussion (I don't go back on my promises) and a happy advisor. Wouldn't life just be easier if you had the steps neatly listed one after the other and all you had to do was carry them out. HAHA! Keep dreaming!

So now down to the scoring of the papers. QMs are supposed to indicate those steps of the protocol that are hard to code up. What happens when the protocol itself is not present where it is supposed to be? I'll leave it for you to decide.

Until next time,
VisualAcuity.

[1] Kortylewski, M. et al. (2009). In vivo delivery of siRNA to immune cells by conjugation to a TLR9 agonist enhances antitumor immune responses. Nature Biotechnology.
[2] Wang, T. et al. (2003). Regulation of the innate and adaptive immune responses by stat-3 signaling in tumor cells. Nature Medicine.
[3] Niu, G. et al. (2002). Constitutive stat3 activity up-regulates vegf expression and tumor angiogenesis. Oncogene.
[4] Turkson, J., Bowman, T., Garcia, R., Caldenhoven, E., Groot, R. P. D., and Jove, R. (1998). Stat3 activation by src induces specific gene regulation and is required for cell transformation. Mol Cell Biol.
[5] Yu, C.-L.,Meyer, D. J., Campbell, G. S., Larner, A. C., Carter-Su, C., Schwartz, J., and Jove, R. (1995). Enhanced dna-binding activity of a stat3-related protein in cells transformed by the src oncoprotein. Science.

Thursday, April 22, 2010

The Emancipation of Bio Protocols

Welcome to biocritic.blogspot.com!
Biologists world-over have put up with the tyranny of bad protocol writers for decades and it is now time for the madness to stop. The posts on this blog will introduce readers to what biologists have to put up with all for the love of the subject (or in some cases, for the sadistic pleasure of their advisors).

The methodology adopted to distinguish the good protocols from the ones that can cause brain damage is this:
  1. Protocols will be chosen at random in the realm of molecular biology. (The target is one bad protocol per day.)
  2. They will be coded up (at least tried to code up) in the language BioCoder (which incidentally is the best (and only) protocol-describing programming language in the world).
  3. Missing steps or information will be denoted by a question mark ("?" aka "QM").
  4. The more QMs ("?", "?", "?".. you get the drift) a protocol scores, the more it indicates that the person trying to code up a protocol feels like a monkey who has just had a lobotomy (read "it is an ambiguous protocol").
So if your protocol reads anything like a page from a _________(your-least-favourite-novel-with-incomplete-plot's name) novel, be scared. Be very scared.
(Note: The previous sentence will receive a penalty of 1 QM because of ambiguity.)

Until next time,
VisualAcuity.