Tagged: #overlyhonestmethods

#overlyhonestmethods – day 4

Twitter bird dead

And that’s the third and last part of #overlyhonestmethods (I love trilogies): (go to part one, and two)
The winners of day 4 are:

Our parsimony analysis was in total disagreement with the Bayesian analysis but who cares anymore 

 

Leftover liquid nitrogen was disposed of by freezing everything for our entertainment. 

 

Image recognition software was coded by my housemate ’cause I knew asking would distract him from his loud girlfriend

 

Room temperature was maintained at 26 degrees celsius, as the research assistant acclimatised to British weather 

 

Our laser source was 650 nM, 250 mW because that’s what we found on eBay. 

 

Prior to euthanasia, each animal was thanked and kissed goodbye. Then I cried 

 

Frames were recorded every 5ps but analyzed every 100ps. It’s a long simulation and I ran out of RAM. 

 

surveys for nocturnal birds were not included as they would have compromised our nightly schedule of drinking and cards

 

The OD600 was taken at 0, 2, 4, 6, and 18 hours because I started the culture too late and fell asleep at my bench. 

 

We used a hierarchical Bayesian analysis in the hope it would intimidate and confuse reviewers. 

 

My advisor added 14 words to the title of my paper so that the university would think I had done something important. 

 

3 replicates were used because 3 is the magical number that makes all results true 

 

I actually already have more precise data, but I’d like to squeeze more than one paper out of this 

 

Variance in dye performance was likely due to 30 year age difference between vials.

 

We microseeded protein crystal trays using hair from a postdoc who left 8 years ago.

 

We used a thesaurus to adapt the introduction from our last 4 papers to a non-plagiarized format 

 

Stimuli for this experiment were inspired by a Monty Python sketch…they worked so I  stuck with it 

 

Samples were conditioned for an extra 12 hours because the lab staff discovered Minecraft. 

 

We used this method of DNA extraction because it meant we got to use a robot.

 

We’re citing Jones et al. because their crap results make ours look good.

 

Limitations: You will never get this particular set of variables to happen again, and we’re not even going to try. 

 

PCR was performed by first praying to the PCR voodoo doll my labmate had fashioned out of pipette tips. 

 

Interrater reliability was 0.95. This really shocked 3rd author, as 1st & 2nd authors can’t even agree on paint colors. 

 

NMR spectra were obtained in deuterated DMF to obscure the residual solvent in the sample. 

 

After heating sealed reaction flask, product was isolated from lab bench surface via paper towel extraction. 

 

Advisor suggested a qualitative approach because “I just can’t handle any more math right now.”

 

We watched a Youtube video on how to calculate the Spearman Rank Correlation for our results.

 

Optimal yields were obtained by weighing in flasks whose exteriors weren’t completely dry after coming off the rotovap.

 

Three ways to get ideal simulation curves: 1. Change input data; 2. Improve platform configuration; 3. Photoshop. 

 

We submitted to this journal because our collaborator is  on the editorial board

 

This is a great idea- it was even better when it was first published in 1968. 

 

By laboratory I mean my kitchen table.

 

Coffee did more for this project than the Principal Investigator. 

 

Ran out of ethanol while in the field. Used cheap vodka instead. 

 

Recruited grad student subjects by putting free pizza in the conference room then locking the door when they came in. 

 

buy your friendly statatician lots of whisky and hand over the data 

 

What we lack in meaningful results we make up for with a witty pun in our title. 

 

Further experiments are out of the scope of this study because I can’t stand my supervisor anymore 

 

phenol-chloroform extraction was used to purify DNA because WE ARE MORE HARDCORE THAN THOU! 

 

“Janitor is listed as co-author for pointing out simple math errors. Cleaning lady contributor for lending bleach.” 

 

The previous postdoc used magic to interpret the results which is why mine are different 

 

inter-rater reliability was .95 because we all had the same tequila-induced hangover. 

 

We have cited a paper that cited another paper that cited another,that cited the original method-we’ve never read it 

 

“Ethanol had to be replaced by methanol due to stress-induced temptation to drink it all.” 

 

We did a netnography of Facebook & Twitter so we could justify our extensive use of social media to the faculty IT team 

 

This algorithm is called like this because we really wanted it to have a funny acronym.

 

It started off as a double blind study. But hey who can keep secrets?!

 

“This result may have several important implications, but we removed those so reviewer #2 would stop arguing.” 

 

The width of our nanodevice is 1500nm because nano sounds cooler than micro. 

 

We chose to study how crickets jump because we wanted to play with the high speed camera…

 

he study was extended while we begged our relatives to wire money to the helicopter company so we could get back.

 

We chose #6, of our 10 working examples, for a detailed qualitative study, because the rest of ’em are bloody boring.

 

We used the low power setting on the sonicator to disperse because the dodgy button wouldn’t turn any higher 

 

The constant mass peak on ms/ms is attributed to volatiles present while the institute’s hallways were being painted. 

 

We thought that “random” and “chance” didn’t sound so good, so “Monte Carlo simulation” it is then 

 

The data were deemed unreliable as 20% of the experimenters were potentially unknowingly under the influence of 

 

We arbitarily increased the errors by 10% because we didn’t believe they could really be that small

 

The terms ‘biodiversity’ and ‘birds’ are used interchangeably throughout this manuscript.

 

The spectrometer was made by a company which no longer exists, in a country which also, no longer exists 

 

a realistic model is too difficult to work with so here’s twenty papers playing with a toy model.

 

In this paper we used pseudo random numbers for the data and pseudy random words for the prose 

 

“quantification was performed blind” I handed my undergrad my images and made him count stuff unrelated to his project 

 

I used students as subjects because rats are expensive and you get too attached to them 

 

This statistical test was chosen because it had the most Greek letters and therefore Science! 

 

The bacterial samples, my lab coat, the bench, my hands, and the floor, were Gram-stained.

 

Expedited ethics review was sought due to impending divorce between primary and secondary supervisors

 

I did the experiment but the PI is a control freak and wrote the paper. I can’t even translate it for you.

 

The dilution fridge can provide temperatures as low as 0.03 degrees above absolute zero. It works primarily by magic.

 

Samples were taken every hour for 72 hr because I live in the lab 

 

The experiment was randomised because I realised I had forgotten to label anything halfway through

 

Alignment was performed by the most sober person in the lab. 

 

Despite not understanding Spanish, I cited this paper based on a Google translate of the abstract

 

the diastereomers are readily separable by chromatography, so long as you use 1kg of silica gel and load at 0.1%. 

#overlyhonestmethods – day 3

Twitter bird dead

day 3 (If you don’t know what I’m talking about, you should check the part 1 here, or If you know it, you are maybe want to go to part 3 here)
The selection of the day:

 

 We don’t know how results were obtained. The postdoc who did all the work has since left to start a bakery.  

The active catalyst was unknown (but the reaction doesn’t work in new glassware)

 

Prior to adding the 2º antibody, Drosophila ovaries were poured on the floor & sucked back up w/ a pipette.

 

Incubation was performed in my bra while biking to collaborators lab, therefore about 37C for 15min

 

The boss puts 3 technicians as co-authors to reduce the impact of the work of the postdoc he hates 

 

You know that meticulous thing I needed to do with all the bacteria cultures? Yeah, I did that like once, just for show 

 

Calculations were performed using the following parameters, because the software crashed when we tried anything else.

 

By “the equipment was modified” we mean “it broke and we used two rolls of duct tape and some aluminum foil.” 

 

We repeated the experiment because the first time I forgot to use an ethanol resistant pen to label the tubes. 

 

PCR is pretty much magic. I wore a hat and yelled at the thermocycler throughout the program

 

You have to sing to the cells, or the transfection won’t work. They like Nick Cave, Leonard Cohen and 80’s pop. 

 

“data not shown” = You do NOT want to see that, and we don’t want to show it! 

 

We’re surprised it worked. The effect of saying ‘touch wood’ during the final step cannot be overestimated 

 

We barely understand what we did. You have no chance, this section is a formality.

 

Time points were spaced sporadically because the PI kept interrupting me. 

 

The PI is last author because they were the last person to read the manuscript before submitting. 

 

python code available on request, but blue screen of death every time I attempt to run it. beware.

 

slides were not coded for blinding before assessing, because I was all alone in the lab. So very, very alone.

 

The first author didn’t write this Methods section and doesn’t understand half of it. 

 

This work was made possible by EPSRC Grant #1234 & eBay from where we scrounged parts to repair our ancient apparatus 

 

Washed with 5 litres of water & extracted with 3 x 200 mL ethyl acetate… because the flask fell off in rotovap bath 

 

Statistical analysis on the data was performed in Excel, as clicking random buttons in SPSS did not work out. 

 

The artifacts in the confocal microscope image in Fig3 are due to sample noise and definitely not a thumbprint 

 

There is a gap in the lightcurve at 2am because the telescope operator looked at porn sites and I was sort of asleep. 

 

Unexpectedly, nicotine was observed in the sample spectrum *cos the postdoc was sneaking a crafty fag next to the GCMS

  

To subtract the background we simply took away the number we first thought of 

 

We used method A because the previous postdoc’s ex PI went to a conference and overheard someone mention it in a bar 

  

If I read the 345 articles I’m citing in my review ? Are you kidding me ? 

 

We chose to pursue the synthesis of macrobicycle 1 because all our other ideas proved un-synthesiseable 

 

Further study is necessary to fully understand the unexpected results shown here. We have sought the help of a priest. 

 

The contribution of Author No.4 is, frankly, a mystery to us; but he’s best mates with the PI, so there he is 

 

To avoid another expt, we counter-argued with referee citing papers from sociology, mediaeval ceramics & wrestling 

 

We modified the standard protocol so that the samples were temporarily transferred to the floor.

 

A combinatorial library was designed based upon what we happened to have in stock… and wasn’t out of date 

 

We’d been drinking when we performed the thought experiment. 

 

Author 1 performed the experiments and wrote the manuscript, Author 2 did some stats, and Author 3 paid for the beer. 

 

We performed a monte carlo simulation as we are big fans of formula 1

 

The completely irrelevant Figure 2 was included to satisfy the reviewer who obviously didn’t understand the paper. 

 

Participants were recruited through coercion 

 

To confirm the findings, we decided to repeat the same experiments in our collaborator’s lab. In Hawaii 

 

A thorough literature search’ = the 1st 5 pages of google results 

 

In-house Perl scripts were generated using my trusty D20 and a table of ASCII codes 

 

The figures were made using paint. 

 

Recrystallizations were done only because I just love using the phrase “mother liquor” as often as possible <hic>

 

Disputes about the tagging of video data were resolved by wrestling rules.

 

Every step in the method was performed ‘like a ninja’ to be sure of extra sciencey awesomeness 

 

Analyzing the results of this $50M grant rests on the shoulders of 1 creepy dude in our basement who knows SQL.

 

My supervisor was too busy to review your paper, so he passed it to me. So I’m doing it. After a night at the pub. 

 

I can’t send you the original data because I don’t remember what my excel file names mean anymore

 

physicists: “We present the solution for the two-electron case; if we cared about more, we’d be chemists instead.

#overlyhonestmethods – day 2

Twitter bird dead

Yesterday night a new hashtag was baptize on twitter: #overlyhonestmethods.
Since then, more than 1800 tweets have been appeared. Mostly funny, some embarrassing, other just scary…
Thousand of scientists were now free to write what they always wanted to write in their publications… the TRUTH… (and some jokes I hope). 
Here a fine selection of my favorites of the first two days (tweeps and hashtags have been removed):
Day 3 online here and 4 here

 

we did experiment 2 because we didn’t know what the fuck to make of experiment 1 

 

incubation lasted three days because this is how long the undergrad forgot the experiment in the fridge 

 

Slices were left in a formaldehyde bath for over 48 hours, because I put them in on Friday and refuse to work weekends. 

 

This project was started to prove someone wrong. They were not. Dammit. 

 

We do it this way bc grad student was trained by a postdoc trained by a grad student who claimed they knew how to do it 

 

This dye was selected because the bottle was within reach 

 

 We kept pushing buttons in our favorite stats software until all our results has stars next to them

 

500 kW laser pulses were employed BECAUSE LASERS. Pew pew. 

 

The tissue was incubated in the haematoxylin bath for 4 mins (or how long our students might need to go for a smoke). 

 

All ultra pure water was passed through 200 micron filters by a reluctant undergraduate project student 

 

PCR conditions are detailed in supp table 1, including optimised underwear colour and superstitious ritual  

 

Experiment was performed as described in X. But that paper isn’t online and I don’t care about you .

 

Our original sample size was 13, but the superstitious lab head made us discard one so we wouldn’t jinx the experiment 

 

A dilution was made somewhere between 1:2000 and 1:500 because our 1ul pipettor is crap 

 

either a 5 minute (RT) or 18 hour (16 degree) ligation protocol was followed, depending on whether I’d eaten that day 

 

Centrifuge until photocopying is done and then incubate for one grande latte’s time 

 

All buffer recipes were obtained by Googling. 

 

Data was replicated after we found out that the student in charge was coming into the lab high the first time around. 

 

the first draft of this manuscript was written in comic sans 

 

Samples were left in the thermocycler at 4°C for 48 hours because I was too hungover to come in. 

 

pH of buffer B was adjusted with HCl… then back a bit with NaOH… then a bit more HCl…  

 

We thank reviewer 3 for suggesting this alternative analysis, but we really don’t mean it. 

 

Software X was run using standard parameters as we have no clue what those flags actually do. 

 

RNA was chilled on ice for 30-minutes because big bang theory was on. 

 

The reaction was complete when the compound changed to green & the crazy Russian postdoc said it tasted ‘OK’ 

 

Statistical analysis performed was a generalised linear mixed effects model because it’s hard and makes us look smart. 


…the chemicals were combined & stirred by hand for 2 hours by our project students as they were getting on our nerves 

 

 Blood samples were spun at 1500rpm because the centrifuge made a scary noise at higher speeds. 

 

Prob is there is an element of truth in every joke…I’m now worried about all the data of  everyone contributing to 

 

“We chose this field site because we’ve got friends in the town and there’s quite a nice pub” 

 

At this point in the protocol I dropped the sample on the floor, but it looked all right so I carried on anyway 

 

“The computer said it, I believe it, that settles it” 

 

We highlighted the chicken’s vascular structure by injecting the femoral artery with PVA, because it amused the prof. 

 

We added glucose to acetaldehyde, for reasons I missed whilst learning japanese numbers from the cute exchange student 

 

An absolutely true  in a protocol someone gave me once: “We added 888 uL because it’s a lucky number in China.”

 

Reagents were mixed in the first vaguely clean looking glass beaker-type thing I could get my hands on 

 

We used a support vector machine because, really, what biologist is going to know what that is? 

 

We decided to use Technique Y because it’s new and sexy, plus hot and cool. And because we could. 

 

Before measurement, samples were kept free from contamination & if we dropped any we totally followed the 5 second rule 

 

It’s wrong. We know it’s wrong. The literature says it’s wrong, but our PI told us to do it so we’re doing it anyway 

 

The sample was diluted in water in a 1:1000 ratio, because the flask fell off the Rotavap  

 

The article was written in passive voice because the PI went through and changed all the active back to passive. 

 

Once PI was banned from lab due to her bad technology-karma, the method and equipment worked without incident 

  

Fig4 is included in the manuscript because it took *ages* to make and it looks really pretty don’t you think? 

 

We’re only citing this paper because one of the authors has a rude/funny last name 

  

Author Contributions. X.Y.Z. provided reagents and was so annoying we had to put him on the paper for his plasmid. 

 

 I didn’t label the tubes, but by looking at the results, I’m pretty sure this one is the control 

 

This experiment followed the method as set out in ref[3], which isn’t available digitally. And is written in Russian. 

 

If a paper’s title is a phrased as a question then the answer is usually no 

 

Glassware was thoroughly cleaned with piranha solution, apart from the top rim as the PhD student was a chicken 

 

Overnight incubations ran for 12-20 hours, depending on how many pubs I hit that evening. 

 

Crystals were obtained by the “I forgot the flask in the fume hood” technique. 

 

Samples 3, 15 and 23 through 41 were discarded due to suspected taco sauce contamination. 

 

We used jargon instead of plain English to prove that a decade of grad school and postdoc made us smart. 


We counted out 2,500 of each snail species. Well, the undergrad did, anyway. Can you believe those guys?  



Day 3 online here and 4 here