Back to blogging and bye bye research

Hello blog world, it’s been a while. My posts are going to change direction from now on as I have left research and am starting a PGCE. I finished my masters last September and have since worked a bit and travelled a bit. It’s been a great year, with not a lot of science, but it has given me time to reflect on my master’s year and why I decided to leave the research world… for now. Completing a masters was a really good experience and I would encourage anybody considering one to go for it (it’s tough, but worth it). But by the end of the year I’d decided that research wasn’t for me. This blog is a bit of an overview about two major problems I came across in scientific research which influenced my decision.

 breaking bad bryan cranston walter white aaron paul jesse pinkman GIFBad Science (thanks Ben (1))

I came across many examples of bad science when reading papers. I’m not going to use actual examples, mainly because this was last year and I’m not prepared to hunt through my old work, so I’m afraid you’ll have to just take my word for it, which you shouldn’t do, because that’s not good science (so maybe just stop reading now?). But here are some examples of bad research practice that I regularly came across:

  1. Conclusions would be drawn from data which didn’t relate to the results. Tenuous links were made and results were skewed to fit with hypotheses.
  2. Some experiments were simply designed badly. For example, basic requirements in an experiment, like controls and repeats (basics taught in year 7 science…) were often not evident.
  3. Using microscopic images to infer things that just weren’t there. One picture of a blurry neurone does not prove anything.
  4. Taking other papers work as given and not exploring how research was completed.

Ploughing through these papers and discussing them in journal clubs made me realise how important ‘good science’ is. But it is hard! I appreciate the inevitability of contrasting results or bizarre outcomes in science, it’s just what happens. That’s why it’s so important to not draw conclusions lightly, but encourage repeats or further experiments. However this does take time and money, which is not always on a scientist’s side. I can almost understand why research groups produce these kinds of papers. The culture in scientific research, both in industry and in universities, is designed around funding. You only have to read Ben Goldacre’s Bad Science(1) and Bad Pharma(2) for some examples of cases where results have been wrongly interpreted, or where research was rushed to meet funding body requirements. The competitiveness of research (often competing for limited funding) can lead to ill-thought out papers and possibly ill-thought out conclusions, which in turn can mislead future studies. It may even prevent progression in a whole area of science. I was lucky to be in a lab where we scrutinised everything in a paper, which meant we were very careful with the work we carried out ourselves. Yes, it was painstaking and led to long days troubleshooting experimental set ups, but it meant we could really trust what we did. However, even in my lab there was a pressure to produce results fast and in time scales that were just not realistic.

Women in Science

Having said all this, there are obviously some excellent researchers out there. I got to meet some of these people at ARUK’s conference in Manchester last year, where I was especially impressed with some of the woman scientists. But it made me realise there is still a long way to go before women in science have equal opportunities. This has already been extensively written about, but I’m going to explain a bit about my experience and perspective. I didn’t experience any sexism during my day-to-day life in research, nor have I spoken to anyone who has. That’s obviously not to say it doesn’t happen, however my experience of gender inequality is more to do with employability. A quick internet search came up with lots research in this area and repeatedly show that women really are disadvantaged when it comes to getting a job in STEM research(3).

One woman I spoke to said she found it nigh-on impossible to find a new post-doc position after having a baby. It was assumed that because of a gap in her publications, she was no longer a desirable post-doc choice as she had been out of science for a year. Now obviously this is just one perspective, but surely there is something inherently wrong when a highly qualified and intelligent woman struggles to find a position because of an assumed lack of up-to-date knowledge? The pressure to perform and have a large and up-to-date back catalogue of publications is a stress for anyone, but especially for women. Any sign of a gap in publications makes it much harder to get back into the profession. Leaving to have a family is not only a problem for women (although more likely to be). The instability of a research career can be a problem for anyone:

Short term contracts are another big problem in science, and not just for women. It is extremely hard to settle down in one place for more than a few years when projects are funded on time scales. Only after years of moving about does it become remotely possible to find a permanent position in a university. This is another reason why women in research struggle. Uprooting every few years is not the way that most people envisage bringing up a family.

This is not to say it is not possible for women. I really admire those women professors who have reached the top of their game and have made it work. Because from my position, it sure doesn’t seem easy to do. If any of you research wonder women are reading this, please let us into the secret!

I’m aware this has been a bit of a rant and perhaps slightly one-sided. Yet after speaking to many people in research, these views are widely shared and are some of the reasons I decided not to pursue a research career. Instead, I’m about to start a PGCE, to teach secondary school biology. Aside from the issues I have with research, I found that the part of my masters I enjoyed the most was outreach and communicating the joys of science to others. But going into a profession where I hope to inspire young women and men to consider a career in science, I am very keen for some of these issues to be addressed. What the answer is, I am not quite sure (another post maybe). Because getting people excited about science is only the first step, keeping them in science is another matter.

  1. Goldacre B. Bad Science. Tantor Media Inc; 2012.
  2. Goldacre B. Bad pharma. London: Fourth Estate; 2013.

Back to basics: DNA – deoxyribo-whaaat

I seem to have a big mix of people reading my blog which is exciting, but I’ve realised it means that some things I talk about might seem straightforward to more scientific readers but pretty confusing for the non-sciencey bunch (i.e. my science noob dad). I’m starting to realise this is one of the biggest difficulties in talking about science, so as well as the current research posts, I’m going to start doing some ‘back to basic’ blogs to help with some of the basic concepts which are fundamental for understanding how our body works.

The first one is going to be DNA (or deoxyribonucleic acid if you’re feeling fancy). As you know, DNA is the mole
cule that determines just about everything about us, from our hair colour to the things that are common to all living organisms like how we reproduce. But how does it do this? This post will try to unpick what it means when we say that DNA ‘codes’ for things and how this links in with genetics.

You’ve probably heard that our DNA is 99% the same as a plant’s DNA. That’s because most of it “codes” for proteins that carry out tasks in our cells that we’re not even aware are being done. It does this by encoding for proteins. When scientists talk about the proteins in our diet that turn us into muscle men, that’s just one tiny portion of the protein in our body. It is the small percentage our body needs that can’t be encoded for by our DNA. Proteins are required for virtually all functions in our cells. I could do hundreds of blogs on proteins, so will leave it at that for now – all we need to remember for this post is that proteins are at the heart of just about everything that happens in our bodies.

DNA is made up of two strands which wrap together to form the famous double helix. These strands contain 3 main parts – phosphate groups, forming the backbone


DNA structure (Image created by Madeleine Price Ball)

of the strands, bases which do the all-important coding (I’ll explain this in a minute) and deoxyribose sugar groups that join the base and the backbone together.

There are four types of bases – adenine, thymine, cytosine and guanine, or A, T, C and G. These pair up with each other to join two strands together to form the double helix.
Each strand is ‘complementary’ to the other, meaning they pair up neatly. A can only join to T while C can only join to G. The order in which these bases appear determines what that particular strand of DNA encodes for.

How does this code fit into genes and chromosomes?
DNA is found in every cell, coiled up into chromosomes. As you probably know, there are 23 pairs of chromosomes (so 46 all together) in humans. Each is made up of around 2000 genes and because there are two copies of each chromosome, there are two copies of every single gene (there is one long strand of DNA in each chromosome and sections of this strand make up the genes – the picture shows this a bit more clearly). I’ll talk in a future post about how these genes are inherited and why certain genes are expressed rather than others.

Now, every cell has the identical full set of chromosomes, but not every cell needs all the DNA to code for every single protein all the time. For example, you don’t need the DNA for the pigment that causes a specific eye colour to be expressed in your big toe. Whole genes (i.e. – whole segments of DNA) are either switched on or off. These switches are controlled by a set of proteins called ‘transcription factors’ – the regulators of DNA expression.


DNA to chromosome (from BBC Bitesize)

DNA to protein
So how does DNA ‘code’ for a protein? DNA is found in the nucleus, whereas proteins are generally required in the rest of the cell – the cytoplasm. The problem is, DNA is too big and bulky to get out of the nucleus to code for proteins directly. That is one of the reasons why there is a step in between. This is where the molecule RNA, or ribonucleic acid, comes in. RNA is very like DNA, but with a couple of differences. For example RNA only has a single strand and uses ribose sugars rather than deoxyribose sugars to bond the bases to the backbone.

So when a certain protein is required, transcription factors (mentioned above) activate the DNA in a specific gene. This DNA undergoes a process called ‘transcription’ – the formation of a new strand of RNA, using the DNA as a template. The RNA contains the exact ‘complimentary’ sequence to DNA. So for example if the DNA sequence was CTGGTC, then the corresponding RNA would be GACCAG (remember how the bases match up). But chromosomes have anywhere between 50,000,000 to 300,000,000 base pairs each, so the strand being sequenced would be a lot longer than 6 letters!

So that’s transcription – the conversion of DNA to RNA in the nucleus of cells. This RNA is smaller and more transportable, so it then moves out of the nucleus through nuclear pores into the cytoplasm, where it can finally start making proteins.

The conversion of RNA to protein is called ‘translation’ – you can think of it as being converted into a whole new ‘language’. This process occurs on a piece of machinery called the ribosome. The ribosome feeds the RNA strand through its ‘reading apparatus’ and ‘reads’ the code. The ribosome can then recruit the correct amino acids required, depending on what it has read on the RNA. Every three bases code for a particular amino acid, depending on the order that they appear. For example, AGT codes for serine while AGA codes for arginine. These 3 letter codes are called codons. As there are only 20 biological amino acids used to make up all proteins, but 64 possible codons (different combinations of the 4 bases to produce a 3 letter code), there is more than one codon that can code for a particular amino acid. And not all 64 codons actually code for an amino acid, some code for a ‘stop’, which signals to the ribosome that the amino acid strand is complete.

Once the amino acid strand has been made, it is transported to various other structures in the cell to be processed into a correctly folded and complete protein.

Things can go wrong at any stage of the process I’ve just described. It requires so many things to go right that it’s not surprising that problems occur from time to time. For example, cancer is often caused by mutated genes, i.e. – mutated DNA sequences. UV light or carcinogens may cause a base pair to be deleted or swapped for another, causing a whole different amino acid to be coded for. Luckily, as you may have realised, our bodies are extremely clever and well adapted to deal with these problems. There are various repair mechanisms to ensure that mutated DNA doesn’t produce misfolded proteins. Often, it is only when these repair mechanisms themselves are mutated that disease occur.

So hopefully, if you’ve made it to the end of this long post, you understand why DNA is so important. It’s quite hard to get your head around how complicated all these tiny processes are. Plus they’re taking place every second in all the billions of cells in your body.