Experiments that I’ve tried and failed

Something that they don’t tell you when you see the cool science headlines is exactly how much time, effort and failure went into getting that exciting paper published. A lot of people that I’ve spoken to, myself included, say that their moods are massively impacted by their results. While it’s easy to say that we shouldn’t let a negative result, or one that doesn’t ‘fit’ with our hypothesis, have such an effect on our mood, it does. It’s very easy to forget that you aren’t the only one faced with disappointment. With that in mind, here are the experiments that I’ve done over the past 6 months that didn’t go exactly…according to plan.

1. Using western blotting to look for differences in protein expression

I am a self professed serial Western blotter. As a molecular biologist, it’s part of my bread and butter, daily protocol, could definitely do it in my sleep. When my Westerns suddenly stopped working, I was not fazed. I checked, I checked again. I ran new gels. I checked that my proteins had transferred correctly. I used new antibodies. I looked at different proteins on the same blot. I was at my wits’ end. If I can’t do this, what kind of scientist am I? I can’t even figure out a problem in a technique I’m pretty much an expert in!


Chocolate. And, it turned out I was using a sample buffer that didn’t have anything to denature proteins in it.

2. Using a viral vector to add a plasmid to my cells

OK, not an expert, I confess. Other people in my lab do this all the time. I’d read the protocols. I’d seen people do it. I was confident. I added my plasmid and my viral vectors. Virus made? Check. Next step? Add the virus to my target cells, and voila, target cells should have the plasmid. Easy, right? When I added the antibiotic to select my plasmid positive cells, I got a HUGE bacterial infection.


Wine. And a deep clean. Also see number 4.

3. Sequencing a CRISPR knockdown

I was feeling pretty good. I’d used CRISPR for the first time to knockdown a protein. I had validated it by western, and it was gone, not a trace could be found. However, it’s good practice to confirm the gene is knocked down by genome sequencing. I ordered primers, checked them against all the appropriate sites. Did the PCR, isolated the band, and sent it for sequencing. Did I get my gene of choice back? No. I got a gene named Pou. I sequenced poo. Literal s**t.


More primers, and accepting that DNA and I do not agree.

4. Making a virus 2.0

After the first failed attempt, I tried again. Same protocol, but with more caution. Changing everything, making sure that I didn’t accidentally touch anything then put it into my flasks. The result? Not a bacterial infection, but a fungal infection!


Hydrogen peroxide bomb, and a new tactic. See number 7.

5. Validating an in vivo experiment by qPCR

After a thesis committee meeting, one of my in vivo experiments was questioned, so we tried to validate using a different diet in normal mice without tumours. I spent a month looking at three different organs and 15 genes by qPCR. Did the qPCR work? Yes! Did it show any difference? Not a dot.


A good night with friends, and realising that no *significant* difference isn’t the end of the world.

6. Tumour sections floating away

One of my main techniques is staining fixed tumours to look at protein expression. Because formalin fixation creates lots of cross-links to preserve the tissue, we have to retrieve antigens that antibodies can bind to before looking for them. For me, this involves boiling the tumour sections in a buffer. What I didn’t expect was to open the press cooker, and see my lovely tumours floating above the slides they used to be on. This wasn’t a one off. This happened 3 times in a row.


Ranty email and getting new sections cut.

7. Adding a plasmid, take 3

I’ve tried it twice, and it failed. Miserably. But this time I had a new tactic. All was going exactly according to plan, which is worrying in itself. Plasmid was in, and cells had been selected with an antibiotic. Time to send them for sorting, because positive cells were expressing GFP. 3 days later, my control plasmid cells come down with a raging infection. I had a beautiful cell culture for my target protein cells, and bacterial soup for my control. Fantastic.


Bleach, lots and lots of bleach. Also starting again and keeping a back up flask.

If I’ve learnt one thing by spending the past 5 years in biology labs, it’s that some of the most successful scientists are also the most stubborn and resilient when faced with constant failure. It can be difficult to remember that things don’t always work, and by trying again, eventually you’ll get THE result.

What were some of your most disheartening science fails?


3 thoughts on “Experiments that I’ve tried and failed

  1. phdfreakingout says:

    I loved this!! especially your solutions including wine, chocolate, friends ect! I try so hard to not let bad results and failed experiments get me down- but weeks of things going tits up gets you in the end! Thank you for a funny and very relatable post πŸ™‚

    Liked by 1 person

  2. Ifigeneia says:

    I remember many if these πŸ˜‚ And I remember discussing my failed experiments with u often as well. Your solutions are gold btw πŸ’•


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s